E V I E W R S April 14 , 2008. (Minor changes may …...E V I E W S I N A D V A N C E...

ANRV345-BI77-25 ARI 29 February 2008 19:15

RE V I E W

S

IN

AD V A

NC

E

Glycosyltransferases:Structures, Functions,and MechanismsL.L. Lairson,1 B. Henrissat,2 G.J. Davies,3

and S.G. Withers1

1Department of Chemistry, University of British Columbia, Vancouver V6T 1Z3,Canada; email: [email protected], [email protected] et Fonction des Macromolecules Biologique, CNRS,Universites Aix-Marseille I and II, Marseille 13288, France;email: [email protected] Biology Laboratory, Department of Chemistry, University of York,Heslington YO10 5YW, United Kingdom; email: [email protected]

Annu. Rev. Biochem. 2008. 77:25.1–25.35

The Annual Review of Biochemistry is online atbiochem.annualreviews.org

This article’s doi:10.1146/annurev.biochem.76.061005.092322

Copyright c© 2008 by Annual Reviews.All rights reserved

0066-4154/08/0707-0001$20.00

Key Words

carbohydrate-modifying enzymes, glycobiology, glycosylation, ionpair mechanisms, nucleophilic substitution

AbstractGlycosyltransferases catalyze glycosidic bond formation using sugardonors containing a nucleoside phosphate or a lipid phosphateleaving group. Only two structural folds, GT-A and GT-B, havebeen identified for the nucleotide sugar-dependent enzymes, butother folds are now appearing for the soluble domains of lipidphosphosugar-dependent glycosyl transferases. Structural and ki-netic studies have provided new insights. Inverting glycosyltrans-ferases utilize a direct displacement SN2-like mechanism involvingan enzymatic base catalyst. Leaving group departure in GT-A foldenzymes is typically facilitated via a coordinated divalent cation,whereas GT-B fold enzymes instead use positively charged sidechains and/or hydroxyls and helix dipoles. The mechanism of retain-ing glycosyltransferases is less clear. The expected two-step double-displacement mechanism is rendered less likely by the lack of con-served architecture in the region where a catalytic nucleophile wouldbe expected. A mechanism involving a short-lived oxocarbenium ionintermediate now seems the most likely, with the leaving phosphateserving as the base.

25.1

Review in Advance first posted online on April 14, 2008. (Minor changes may still occur before final publication online and in print.)

Ann

u. R

ev. B

ioch

em. 2

008.

77. D

ownl

oade

d fr

om a

rjou

rnal

s.an

nual

revi

ews.

org

by U

NIW

ER

SYT

ET

WR

OC

LA

WSK

I on

04/

18/0

8. F

or p

erso

nal u

se o

nly.


Contents

INTRODUCTION. . . . . . . . . . . . . . . . . 25.2Glycosyltransferase Activities . . . . . 25.2Glycosyltransferase Folds . . . . . . . . . 25.2Glycosyltransferase

Classification . . . . . . . . . . . . . . . . . . 25.5Genomic Distributions of

Glycosyltransferases . . . . . . . . . . . 25.6Mechanism of Inverting

Glycosyltransferases . . . . . . . . . . . 25.7Inverting GT-A

Glycosyltransferases . . . . . . . . . . . 25.7Inverting GT-B

Glycosyltransferases . . . . . . . . . . .25.10Mechanism of Retaining

Glycosidases . . . . . . . . . . . . . . . . . .25.12Mechanism of Retaining

Glycosyltransferases . . . . . . . . . . .25.13Challenges in Studying the

Mechanisms of RetainingGlycosyltransferases . . . . . . . . . . .25.13

Retaining GT-AGlycosyltransferases . . . . . . . . . . .25.15

Retaining GT-BGlycosyltransferases . . . . . . . . . . .25.20

An Alternative SNi-LikeMechanism . . . . . . . . . . . . . . . . . . .25.23

Evolutionary Constraints uponRetaining GlycosyltransferaseMechanisms. . . . . . . . . . . . . . . . . . .25.24

Concluding Remarks on theMechanism of RetainingGlycosyltransferases . . . . . . . . . . .25.27

INTRODUCTION

The enormous complexity of the variousoligosaccharide structures found in nature (1)is derived from a rational orchestration ofthe enzymatic formation and the breakdownof glycosidic linkages achieved by glycosyl-transferases, glycosidases, glycan phosphory-lases, and polysaccharide lyases. In contrastto the well-characterized mechanistic strate-gies used by glycosidases to catalyze glyco-

sidic bond hydrolysis (2, 3), the mechanisms ofthe glycosyltransferases responsible for glyco-side bond formation remain less clear. Despitea lack of evolutionary relatedness, glycosyl-transferases had been thought to use mech-anistic strategies that directly parallel thoseused by glycosidases (2). However, some dis-tinct differences are becoming apparent, asdiscussed below.

Glycosyltransferase Activities

Glycosyltransferases are most accurately de-fined as those enzymes that utilize an acti-vated donor sugar substrate that contains a(substituted) phosphate leaving group. Donorsugar substrates are most commonly acti-vated in the form of nucleoside diphosphatesugars (e.g., UDP Gal, GDP Man); how-ever, nucleoside monophosphate sugars (e.g.,CMP NeuAc), lipid phosphates (e.g., dolicholphosphate oligosaccharides), and unsubsti-tuted phosphate are also used. Nucleotidesugar-dependent glycosyltransferases are of-ten referred to as Leloir enzymes, in honorof Luis F. Leloir who discovered the firstsugar nucleotide and was awarded the NobelPrize in chemistry in 1970 for his enormouscontributions to our understanding of glyco-side biosynthesis and sugar metabolism. Theacceptor substrates utilized by glycosyltrans-ferases are most commonly other sugars butcan also be a lipid, protein, nucleic acid, an-tibiotic, or other small molecules. In addition,although glycosyl transfer most frequently oc-curs to the nucleophilic oxygen of a hydroxylsubstituent of the acceptor, it can also occurto nitrogen (e.g., the formation of N-linkedglycoproteins), sulfur (e.g., the formation ofthioglycosides in plants), and carbon (e.g., C-glycoside antibiotics) nucleophiles.

Glycosyltransferase Folds

As has been done for other classes ofcarbohydrate-active enzymes, glycosyltrans-ferases are classified into families basedon amino acid sequence similarities (4, 5).

25.2 Lairson et al.

Ann

u. R

ev. B

ioch

em. 2

008.

77. D

ownl

oade

d fr

om a

rjou

rnal

s.an

nual

revi

ews.

org

by U

NIW

ER

SYT

ET

WR

OC

LA

WSK

I on

04/

18/0

8. F

or p

erso

nal u

se o

nly.


Indicative of the pace and progress of genomesequencing endeavors, in the 10 years sinceCampbell and colleagues first started sucha classification of glycosyltransferases, thenumber of distinct families has grown from27 to 90.

There are marked contrasts between thediversity of three-dimensional (3-D) foldsobserved for hydrolytic glycoside hydrolasesand the very limited folds used by the syn-thetic glycosyltransferases. Structural char-acterization of representatives from a largenumber of the 110 families of glycosidases(http://www.cazy.org) has revealed an ex-traordinary degree of diversity in overall fold,despite considerable commonality in active-site features. This would indicate a conver-gence during the evolution of catalytic mecha-nism. In contrast, the recent burst of reportedglycosyltransferase structures has revealed aquite different situation as only two generalfolds, called GT-A and GT-B (nomencla-ture proposed in Reference 6), have been ob-served for all structures of nucleotide-sugar-dependent glycosyltransferases solved to date(5, 7, 8). Furthermore, threading analysis hasrevealed that many of the uncharacterizedfamilies are also predicted to adopt one ofthese two folds. This finding may indicatethat the majority of glycosyltransferases have

GT-A fold:glycosyltransferaseprotein topologyconsisting of twoclosely abuttingβ/α/β Rossmanndomains

GT-B fold:glycosyltransferaseprotein topologyconsisting of twoβ/α/β Rossmanndomains that faceeach other and arelinked flexibly

evolved from a small number of progenitorsequences. However, it also reflects the re-quirement for at least one nucleotide-bindingdomain of the Rossmann fold type. Tantaliz-ing support for the former notion is derivedfrom the fact that only two glycosyltransferasefamilies (GT2 having the GT-A fold and GT4having the GT-B fold) are possessed by prim-itive Archae, which may reflect the originsfrom which the vast majority of glycosyltrans-ferases have evolved (5). In addition, withinthe GT4 family, there exist members that uti-lize donor substrates activated with a nucleo-side diphosphate and others that utilize donorsubstrates activated with a phosphate group,suggesting an evolutionary link between en-zymes that utilize these two substrate forms.

The GT-A fold was first described for theinverting enzyme SpsA from Bacillus subtilis,for which both apo- and UDP-bound 3-D X-ray crystal structures were obtained (9). Con-sisting of an open twisted β-sheet surroundedby α-helices on both sides, the overall archi-tecture of the GT-A fold is reminiscent oftwo abutting Rossmann-like folds, typical ofnucleotide-binding proteins. Two tightly as-sociated β/α/β domains, the sizes of whichvary, abut closely, leading to the formationof a continuous central β-sheet (Figure 1a).For this reason, some describe the GT-A

a GT-A b GT-B

Figure 1Overall folds observed for glycosyltransferase enzymes. (a) The GT-A fold is represented by theinverting enzyme SpsA from Bacillus subtilus, Protein Data Bank (pdb) 1qgq, and (b) the GT-B fold, bybacteriophage T4 β-glucosyltransferase, pdb 1jg7.

www.annualreviews.org • Glycosyltransferases 25.3

Ann

u. R

ev. B

ioch

em. 2

008.

77. D

ownl

oade

d fr

om a

rjou

rnal

s.an

nual

revi

ews.

org

by U

NIW

ER

SYT

ET

WR

OC

LA

WSK

I on

04/

18/0

8. F

or p

erso

nal u

se o

nly.


GT-C fold: apredicted proteintopology fortransmembraneglycosyltransferasesthat is notexperimentallyverified

fold as a single domain fold. However, dis-tinct nucleotide- and acceptor-binding do-mains are present (7). Eukaryotic memberspossessing the GT-A fold typically have ashort N-terminal cytoplasmic domain that isfollowed by transmembrane and stem regionsleading to the globular catalytic region (10).It should be noted that not all enzymes thatpossess a GT-A fold are glycosyltransferases.For example, and possibly indicating a diver-gence in mechanism during the course of evo-lution, the sugar-1-phosphate pyrophospho-rylase/nucleotidyl transferase superfamily ofenzymes, responsible for the synthesis of nu-cleoside diphosphate sugars, has been shownto possess the GT-A architecture (11). Poten-tial confusion arises with enzymes that almostdisplay the canonical GT-A topology, but witha different order of β-strands. This is exempli-fied by sialyltransferases from family GT-42(12, 13) whose structure could either be con-sidered a new fold type (on the basis of differ-ing connectivity) or as a modified GT-A fold(a description the original authors favor) (12).

Most GT-A enzymes possess an Asp-X-Asp (referred to as DXD) signature inwhich the carboxylates coordinate a divalentcation and/or a ribose (14, 15). These areby no means conserved motifs (none of theresidues is invariant), and although frequentlydescribed as a determining characteristic ofGT-A glycosyltransferases, examples do nowexist of enzymes from this fold family that donot possess the DXD “signature” (16). Fur-thermore, many sequences possess a DXDsignature, but are not glycosyltransferases.

The first determined 3-D structure fora nucleoside diphosphate-utilizing glycosyl-transferase was in fact reported in 1994and was that of a DNA-modifying β-glucosyltransferase from bacteriophage T4(17). The overall fold of this protein was foundto be homologous to that of glycogen phos-phorylase, and now that it has been observedfor other glycosyltransferases, it is termed theGT-B fold. Like the GT-A fold, the architec-ture of GT-B enzymes consists of two β/α/βRossmann-like domains; however, in this case,

the two domains are less tightly associatedand face each other with the active-site ly-ing within the resulting cleft (Figure 1b). Aswith the GT-A fold, these domains are asso-ciated with the donor and acceptor substrate-binding sites. Non-glycosyltransferase en-zymes are also known to adopt the GT-B fold,with UDP GlcNAc 2-epimerase serving asone such example (18).

A third glycosyltransferase fold termedGT-C was recently predicted on the basis ofiterative sequence searches, using programssuch as BLAST, (19). This report predictedthat eight families (GT22, GT39, GT48,GT50, GT53, GT57, GT58, GT59) wouldpossess the GT-C fold. Following the publi-cation of this report, 4 (GT66, GT83, GT85,GT86) of the 15 glycosyltransferase fami-lies subsequently created were also predictedto adopt this fold by the CAZY database(http://www.cazy.org). The predicted archi-tecture of the GT-C fold is that of a large hy-drophobic integral membrane protein locatedin the endoplasmic reticulum or on the plasmamembrane having between 8 and 13 trans-membrane helices and an active site located ona long-loop region (20–22). Consistent withthis predicted alternative architecture is thefact that 10 of the 12 families (all but GT48and GT53) predicted to adopt the GT-C foldutilize lipid phosphate-activated donor sugarsubstrates. A subsequent comparison of gly-cosyltransferase families using a “profile hid-den Markov method” of sequence analysis re-sulted in classification of family GT48, whichuses UDP Glc as donor substrate, within theGT-A superfamily (23). Members of familyGT53, predicted to adopt the GT-C fold, uti-lize UDP L-arabinose as a donor substrate.

Very recently the first 3D structure of anenzyme assigned to GT-C was determined:the soluble C-terminal domain of the Pyrococ-cus furiosius oligosaccharyltransferase STT3(23a). This structure weakens the value of theGT-C classification because it reveals that theportion of the protein on which the alignmentis based is the trans-membrane region, whichdoes not include the predicted loop bearing

25.4 Lairson et al.

Ann

u. R

ev. B

ioch

em. 2

008.

77. D

ownl

oade

d fr

om a

rjou

rnal

s.an

nual

revi

ews.

org

by U

NIW

ER

SYT

ET

WR

OC

LA

WSK

I on

04/

18/0

8. F

or p

erso

nal u

se o

nly.


the active site. Rather, the active site seemsto be located within this soluble C-terminaldomain, though a note of caution is necessaryas the authors were not able to demonstratecatalytic activity for this truncated protein.However, this inactivity could well be a con-sequence of the need for the transmembraneregion to bind the lipid portion of the sub-strates. Consequently, even though the mem-bers of GT-C are likely related, it is prob-ably through their trans-membrane regionrather than through their catalytic domain.It is therefore quite possible that other foldscould be found for these catalytic domainswhose structures are not constrained by theneed to bind a nucleotide; thus, it is unlikelythey will bear Rossmann folds. The soluble C-terminal domain of the Pyrococcus OST- STT3in fact adopts a novel architecture with a cen-tral, mainly α-helical domain surrounded bythree β-sheet-rich domains.

The structure of a peptidoglycan syn-thesizing glycosyl transferase (GT51), an-other enzyme that uses a lipid phosphosugardonor, was also recently determined, as dis-cussed in greater detail below (25, 26). Fas-cinatingly, this enzyme was found to have abacteriophage-lysozyme-like fold, and not aRossmann fold, consistent with it not being anucleotide phosphosugar-dependent enzyme.It was not a predicted GT-C member. Othersequence families, such as GT76, are cur-rently orphan families that are not predictedto adopt the GT-A, GT-B, or proposed GT-Cfold.

Glycoside hydrolases, phosphorylases,and glycosyltransferases all catalyze glycosylgroup transfer. This does lead to confusionand inconsistency in the nomenclature forthese enzymes and also obscures the place-ment of some of these enzymes into CAZyfamilies. For example, the structural andmechanistic study of chitobiose phospho-rylase from Vibrio proteolyticus revealed thatenzymes from family GT36 share more of astructural and evolutionary relationship withglycosidases of clan GH-L, which have an(α/α)6 fold (24). As a result, this family was

subsequently reclassified as family GH-94.Very recently, however, the much-anticipated3-D X-ray crystal structure of a peptidogly-can glycosyltransferase from family GT51was solved with a bound substrate analogue/inhibitor (25, 26). As mentioned above,members of this inverting family utilize lipidII phosphate-activated donor substrates,and the structure of the glycosyltransferasedomain was not found to show similarity tothe GT-A, GT-B, or proposed GT-C folds.Instead, the α-helical fold and several active-site features are more akin to the lysozymefamily of enzymes. Given that peptidoglycanglycosyltransferases are solely synthetic, itwould be questionable to reclassify them asglycoside hydrolases, but given their strongsimilarities to some lysozymes, they certainlyhighlight the muddy waters of classification.Future results with the other orphan familiescurrently classified as glycosyltransferases,but not predicted by sequence analysisto adopt either of the GT-A, GT-B, orproposed GT-C folds, will undoubtedlyshed more light on the evolutionary originof glycosyltransferase activities and theirrelationships to the other major classes ofcarbohydrate-active enzymes. Whether thesefamilies will be found experimentally toactually adopt one of the GT-A, GT-B, orlipid phosphosugar glycosyltransferase foldsor possibly to form new glycosyltransferasesuperfamilies or simply succumb to reclas-sification, as described above, remains to bedetermined.

GlycosyltransferaseClassification

Two stereochemical outcomes are possible forreactions that result in the formation of a newglycosidic bond: the anomeric configurationof the product can either be retained or in-verted with respect to the donor substrate(Figure 2). As such, enzymes catalyzing gly-cosyl group transfer are classified as either in-verting or retaining, depending on the out-come of the reaction. Logically, it follows that


Ann

u. R

ev. B

ioch

em. 2

008.

77. D

ownl

oade

d fr

om a

rjou

rnal

s.an

nual

revi

ews.

org

by U

NIW

ER

SYT

ET

WR

OC

LA

WSK

I on

04/

18/0

8. F

or p

erso

nal u

se o

nly.


Figure 2Like glycosidases,glycosyltrans-ferases catalyzeglycosyl grouptransfer with eitherinversion orretention of theanomericstereochemistrywith respect to thedonor sugar (5).

Invertingglycosyltransferase:a glycosyltransferasethat catalyzes grouptransfer reactionswith net inversion ofstereochemistry atthe anomericreaction center ofthe donor substrate

Retainingglycosyltransferase:a glycosyltransferasethat catalyzes grouptransfer reactionswith net retention ofstereochemistry atthe anomericreaction center ofthe donor substrate

this stereochemical outcome must result fromthe utilization of different mechanisms by thetwo classes of enzymes.

In the case of glycoside hydrolases, withonly very few exceptions (family GH4 en-zymes, goose lysozymes, and soluble lytictransglycosylases from family GH23), the en-zymes within a given GH family catalyze hy-drolysis with the same stereoselectivity. Thisshould also be the case among glycosyltrans-ferase families. However, given the highersequence similarity within the nucleotide-binding domains, which risks dominating thesequence searches and is common to enzymescatalyzing both mechanisms, it is possible thatsome families may contain enzymes that carryout reactions with different stereochemicaloutcomes. Nevertheless, glycosyltransferasefamilies can be classified into clans dependingon their fold and the stereochemical outcomeof the reactions that they catalyze (Figure 3).Among GT-A and GT-B superfamilies, theoverall fold of the enzyme does not dictatethe stereochemical outcome of the reactionthat it catalyzes, as examples of both invertingand retaining glycosyltransferases have beenidentified within both the GT-A and GT-Bfold classes (5). Indicative of this phenomenonare recent findings from studies of a manno-sylglycerate synthetase, which transfers man-nose to the 2-OH of D-lactate, D-glycerateor glycollate with net retention of configu-

ration (27). Based on amino acid sequence,this enzyme was initially classified among theGT-A inverting family GT2, however, struc-tural and mechanistic studies led to its reclas-sification among the GT-A retaining familyGT78. To date, all enzymes predicted to adoptthe GT-C fold belong to inverting glycosyl-transferase families.

Genomic Distributionsof Glycosyltransferases

Examination of more than 500 completelysequenced organisms, listed in the CAZydatabase (see related resource) as of Septem-ber 2007, reveals that by and large the numberof GTs encoded by the genome of an organ-ism correlates reasonably well with the totalnumber of genes of the genome. GTs accountfor about 1% to 2% of the gene products of anorganism, whether archaeal, bacterial, or eu-karyotic. This proportion seems to hold trueeven for double-stranded DNA viruses. Thepopulations of the various GT families alsoshow wide variations in the CAZy database,with two large families, GT2 and GT4, ac-counting for about half of the total num-ber of GTs. Thus organisms, such as plants,with very large genomes that synthesize acomplex cell wall or use the glycosylation ofsmall molecules to tune bioactivity have manyGTs (Arabidopsis encodes approximately 450

25.6 Lairson et al.

Ann

u. R

ev. B

ioch

em. 2

008.

77. D

ownl

oade

d fr

om a

rjou

rnal

s.an

nual

revi

ews.

org

by U

NIW

ER

SYT

ET

WR

OC

LA

WSK

I on

04/

18/0

8. F

or p

erso

nal u

se o

nly.


glycosyltransferases and Populus more than800). In contrast, organisms that have under-gone massive gene loss during evolution tobecome obligate symbionts or obligate par-asites appear to have very few or sometimesno detectable glycosyltransferase genes at all(e.g., several Mycoplasma species). The mas-sive number of plant glycosyltransferases isdue to the presence of several extremely pop-ulated glycosyltransferase families. For exam-ple, Arabidopsis, Oryza, and Populus have re-spectively ∼120, ∼200, and ∼300 family GT2genes. One of the surprises that came withthe completion of the first animal genomeswas that humans have slightly less glycosyl-transferase genes (∼230) than the nematodeCaenorhabditis elegans (∼240).

Mechanism of InvertingGlycosyltransferases

Like inverting glycoside hydrolases, themechanistic strategy employed by invertingglycosyltransferases is that of a direct displace-ment SN2-like reaction. An active-site sidechain serves as a base catalyst that deproto-nates the incoming nucleophile of the accep-tor, facilitating direct SN2-like displacementof the activated (substituted) phosphate leav-ing group (Figure 4a). The key questionsin examining the catalytic mechanism of in-verting glycosyltransferases are, therefore, theidentity of the base catalyst and the methodused to facilitate departure of the (substituted)phosphate leaving group.

Inverting GT-A Glycosyltransferases

As mentioned above, the family GT2 en-zyme SpsA from B. subtilis was the first glyco-syltransferase experimentally shown to adoptthe GT-A fold (9). This family representsthe largest and evolutionarily most ancient ofinverting glycosyltransferases. Unfortunately,the natural donor and acceptor sugars ofSpsA are not known. However, fortuitously,a molecule of the cryoprotectant glycerol wasfound bound within the SpsA active site in

GT-A

GT-B

Non-GT

Retaining - clan IV

Inverting - clan II

Non-GT

Retaining - clan III6, 8, 15, 24, 27, 34*,44, 45*, 55, 60*, 62*,64, 78, 81

2, 7, 12, 13, 14, 16*,21, 25*, 31, 40*, 42,43, 49*, 82, 84

1, 9, 10, 17*, 19, 23,26*, 28, 30, 33, 41,47*, 56*, 63, 80

3, 4, 5, 20, 32*, 35,72

Inverting - clan I

Figure 3Glycosyltransferase (GT) classification system proposed by Coutinho et al.(5). Families are classified into clans on the basis of their fold and activity.GT family numbers belonging to each clan are indicated on the far right.Bona fide families having members with solved 3-D structures are indicatedin red. The remaining families are those predicted to adopt either the GT-Aor GT-B fold. Families identified in black with an asterisk are those withstructures predicted to adopt either the GT-A or GT-B fold solely by Liu &Mushegian (19), and those in black without an asterisk have GT-A or GT-Bstructures as predicted by both Liu & Mushegian and the CAZY Web site.This classification system does not include 39 of the 90 glycosyltransferases.Members from 12 (GT22, GT39, GT48, GT50, GT53, GT57, GT58,GT59, GT66, GT83, GT85, GT86) of those families not included werepredicted to adopt a proposed GT-C fold. On the basis of a determined3-D structure, family GT36 was reclassified among the glycosidases asfamily GH94. Structural characterization of the remaining 26 orphanfamilies (GT11, GT18, GT29, GT37, GT38, GT46, GT51, GT52,GT54, GT61, GT65, GT67, GT68, GT69, GT70, GT71, GT73, GT74,GT75, GT76, GT77, GT79, GT87, GT88, GT89, GT90) will provideinsights into the strengths and limitations of predictive bioinformatic tools.

a suitable position to mimic the natural ac-ceptor sugar and hydrogen bonded to theside chain carboxylate of Asp191, leadingto the proposition that this residue playedthe role of base catalyst in the proposed di-rect displacement mechanism (Figure 5a) (9).This notion was later supported by super-positioning (28) the SpsA structure on thesolved GT-A structures of the enzymes lac-tose synthase (Gal-T1) from family GT7 (29–35), GnT-I from family GT13 (36, 37), andGlcAT-I from family GT43 (38, 39). Theseenzymes all have a conserved Asp or Gluresidue within their active sites, occupyinga position equivalent to that of Asp191 inSpsA (Figure 5). Similarly, an analogouslypositioned conserved Asp or Glu residue has


Ann

u. R

ev. B

ioch

em. 2

008.

77. D

ownl

oade

d fr

om a

rjou

rnal

s.an

nual

revi

ews.

org

by U

NIW

ER

SYT

ET

WR

OC

LA

WSK

I on

04/

18/0

8. F

or p

erso

nal u

se o

nly.


Figure 4(a) Inverting glycosyltransferases utilize a direct displacement SN2-like reaction that results in aninverted anomeric configuration via a single oxocarbenium ion-like transition state. (b) The proposeddouble-displacement mechanism for retaining glycosyltransferases involves the formation of a covalentlybound glycosyl-enzyme intermediate. Abbreviations: R, a nucleoside, a nucleoside monophosphate, alipid phosphate, or phosphate (phosphorylases classified as glycosyltransferases); and R’OH, an acceptorgroup (e.g., another sugar, a protein, or an antibiotic).

been observed in the subsequently obtainedstructures of the enzymes C2GnT-L fromfamily GT14 (16), Mfng from family GT31(40), and GlcAT-P from family GT43 (41, 42)(Figure 5). Structures with bound acceptorsubstrates indicate that these conserved car-boxylates are within hydrogen-bonding dis-tance of the nucleophilic hydroxyl under-

going reaction, consistent with their rolesas base catalysts (Figure 5). Further supportcomes from site-directed mutagenesis studiesof the family GT2 enzyme ExoM in whichit was shown that mutation of Asp187 (struc-turally homologous to Asp191 in SpsA) abol-ished all in vitro glycosyltransferase activity(43).

25.8 Lairson et al.

Ann

u. R

ev. B

ioch

em. 2

008.

77. D

ownl

oade

d fr

om a

rjou

rnal

s.an

nual

revi

ews.

org

by U

NIW

ER

SYT

ET

WR

OC

LA

WSK

I on

04/

18/0

8. F

or p

erso

nal u

se o

nly.


a SpsA b Gal-Tl

c

e

GnTl d Mfng

GlcAT-l f GlcAT-P

UDP

UDP GlcNAc

UDP

UDP

UDPUDP

D255

D318

Glucose

D254

D191

Glycerol

D99

D291

D142

D194D195

D284

Gal

GlcNAc

D197

D281

D196

Gal

Gal

D144

D232

E211

D213

D98

(pdb 1qgq)(GT2)

(pdb 2fyd)(GT7)

(pdb 1foa)(GT13)

(pdb 2j0b)(GT31)

(pdb 1fgg)(GT43)

(pdb 1v84)(GT43)

Figure 5Comparison of the active sites of several metal-dependent inverting GT-A fold glycosyltransferases. Aconserved side chain carboxylate is located in a near identical relative position within all active sites onthe β-face of the donor substrate and (when present) within hydrogen-bonding distance of thenucleophilic hydroxyl of the acceptor sugar and plays the role of the base catalyst in a direct displacementmechanism. Mn2+ cations are shown as purple spheres.


Ann

u. R

ev. B

ioch

em. 2

008.

77. D

ownl

oade

d fr

om a

rjou

rnal

s.an

nual

revi

ews.

org

by U

NIW

ER

SYT

ET

WR

OC

LA

WSK

I on

04/

18/0

8. F

or p

erso

nal u

se o

nly.


The vast majority of enzymes from glyco-syltransferase Clan I that have been subjectedto biochemical analysis use an essential diva-lent cation (usually Mn2+ or Mg2+), coordi-nated by the so-called DXD motif, to facili-tate departure of the nucleoside diphosphateleaving group by electrostatically stabilizingthe developing negative charge. Exceptions tothis strategy have been reported for the metalion-independent sialyltransferases from fam-ily GT42 (12) and β-1,6-GlcNAc transferaseC2GnT-L from family GT14 (16), which usetyrosyl hydroxyls or basic amino acids, re-spectively, to electrostatically stabilize substi-tuted phosphate leaving groups. This strategyis reminiscent of that used by metal ion-independent glycosyltransferases possessing aGT-B fold (as discussed below), indicative of aconvergence of mechanisms among these twosuperfamilies. A convergence in mechanismsbetween these two superfamilies is also illus-trated by the finding that the most likely can-didate for the base catalyst in family 42 sialyl-transferases, as revealed by crystal structureswith a bound CMP 3F NeuAc donor substrateanalogue (12, 13), is a His residue, as is also thecase for many GT-B fold-inverting enzymes(as discussed below).

Theoretical support for a concerted SN2-like displacement mechanism for invertingglycosyltransferases from Clan I has beenprovided by a hybrid quantum mechanical/molecular mechanical study of the β-1,2-GlcNAc transferase GnT-I (44). The resultssupported a catalytic mechanism involving aconcerted SN2-type transition state involvinga near simultaneous nucleophilic attack, fa-cilitated by proton transfer to the catalyticbase (Asp291), and leaving group dissociationsteps. An activation energy of ∼19 kcal/molwas estimated for the proposed transitionstate.

Inverting GT-A glycosyltransferases, no-tably those in family GT2, are involved inthe formation of many β-linked polysaccha-rides, such as cellulose, chitin, and hyaluro-nan. These enzymes exemplify a continuingcontroversy in the polysaccharide field: Is the

UDP monosaccharide the donor (in whichcase the growing chain extends by additionto its nonreducing end) or is a UDP-growingchain the donor with a UDP monosaccha-ride acting as the acceptor (which would re-sult in a growing chain extending by additionat its reducing end)? Labeling experiments oncellulose synthase (45) support nonreducing-end elongation, as does direct analysis of boththe chitooligosaccharide synthase Nod fac-tor C and zebra fish chitin synthase DG42(46). Furthermore, yeast chitin synthases havebeen shown to not use UDP chitobiose as adonor (47). Family GT2 hyaluronan synthasesof class II also clearly use UDP monosaccha-rides (48) as the donor, but class I hyaluronansynthases, closely related to the chitin syn-thases mentioned above, have conversely beenproposed to use reducing-end addition (49).The area is clearly confusing as may be thecase with any polysaccharide synthases [in-deed such controversy also stalked the pep-tidoglycan synthase field until the recentlyreported GT51 structure determination(25, 26)].

Inverting GT-B Glycosyltransferases

Despite sharing homologous 3-D architec-tures, members of this superfamily seemto display a greater degree of diversity inthe selected modes of catalyzing glycosylgroup transfer between, and in some caseseven among families, compared to whathas been observed for inverting GT-A en-zymes. This greater diversity in mechanismis perhaps a reflection of the greater diversityof chemistries catalyzed by this superfamilyof enzymes and perhaps also by the physicalseparation of the two domains.

The prototype of this superfamily is theβ-glucosyltransferase (BGT) from T4 bac-teriophage, the first nucleoside diphosphate-utilizing glycosyltransferase for which a 3-DX-ray crystal structure was obtained (17). Thisunique enzyme, the sole member of familyGT63, transfers glucose from UDP Glc tothe hydroxymethyl substituents of modified

25.10 Lairson et al.

Ann

u. R

ev. B

ioch

em. 2

008.

77. D

ownl

oade

d fr

om a

rjou

rnal

s.an

nual

revi

ews.

org

by U

NIW

ER

SYT

ET

WR

OC

LA

WSK

I on

04/

18/0

8. F

or p

erso

nal u

se o

nly.


cytosine bases. This process has evolved as adefense mechanism to prevent genome degra-dation by phage and host nucleases (50). Inaddition to the initially reported structureswith bound metal ions and UDP present(17, 51), a subsequent ternary complex X-ray crystal structure with bound UDP and aDNA acceptor was obtained (52). The ternarycomplex structure revealed the ability of thisglycosyltransferase to facilitate selective gly-cosylation by inducing the DNA to “flip out”in a fashion reminiscent of DNA glycosy-lases, methyltransferases, and endonucleases.Analogous to what has been found with theGT-A superfamily, mutagenesis studies re-vealed Asp100 to be a likely candidate for abase catalyst. This is supported by the ob-servation that cocrystallization of the wild-type enzyme with UDP Glc results in com-plete hydrolysis of the donor sugar, yieldingUDP product complexes, whereas analogouscocrystallization procedures with the D100Amutant resulted in the observation of intactdonor sugar bound within the active site (53).A crystal structure with bound intact donorsubstrate was also obtained by a brief soak-ing of wild-type BGT crystals with UDP Glc.Interestingly, this donor substrate complexhad no metal cation bound within the ac-tive site despite being soaked with high con-centrations of Mg2+ (53). Instead, positivelycharged side chains were found to neutral-ize the negative charges of the pyrophosphategroup of the bound donor substrate. In con-trast, in the UDP product complex, obtainedby cocrystallization with Mn2+ and UDP, anMn2+ cation was found coordinating the py-rophosphate group and was located in the re-gion occupied by the glucose moiety in theUDP Glc-complexed structure. This led theauthors to propose that the divalent cationplays a role in facilitating product release andnot necessarily in the cleavage of the glyco-sidic linkage of the donor substrate. This al-ternative mode of leaving group activation,compared to the GT-A superfamily, has alsobeen proposed for several other GT-B en-zymes (described below).

Glycosyltransferases from family GT1adopt the GT-B fold and are responsiblefor the glycosylation of various importantorganic structures, such as terpenes, antho-cyanins, cofactors, steroids, peptide antibi-otics, and macrolides, making this one ofthe most intensely studied families of glyco-syltransferases. Three family GT1 enzymes[GtfA (54), GtfB (55), and GtfD (56)] involvedin the biosynthesis of the peptide antibioticvancomycin have been subjected to structuraland biochemical analysis. The results of thesestudies suggest that these enzymes use an as-partate as the catalytic base and that leavinggroup departure is facilitated in a metal ion-independent fashion using a helix dipole andinteractions with side chain hydroxyl and imi-dazole groups to stabilize the developing neg-ative charge. A second group of related en-zymes from family GT1 have also been char-acterized, revealing interesting differencesfrom the Gtf enzymes. These include the mul-tifunctional terpene/flavonoid glycosyltrans-ferase UGT71G1 (57), the flavonoid glu-cosyltransferase VvGTI (58), the macrolideglycosyltransferases OleD and OleI (59), thehuman drug-metabolizing glucuronyltrans-ferase UGT2B7 (60) and the bifunctional N-and O-glucosyltransferase from Arabidopsisthaliana (60a). Althoughthese enzymes werefound to use a metal ion-independent mecha-nism analogous to that described above to fa-cilitate leaving group departure, the catalyticbase was determined to be a side chain imida-zole that interacts with an adjacent conservedside chain carboxylate group. These conclu-sions were based on comparisons to a 3-DX-ray crystal structure with a bound accep-tor substrate (58), whereby the putative sidechain imidazole was found within hydrogen-bonding distance of the reactive acceptor sub-strate hydroxyl and confirmed by kinetic anal-ysis of site-directed mutants.

Other inverting GT-B enzymes that havebeen subjected to structural and mechanis-tic analysis include the heptosyltransferaseWaaC from family GT9 (61), the fucosyl-transferases FucT from family GT10 (62)


Ann

u. R

ev. B

ioch

em. 2

008.

77. D

ownl

oade

d fr

om a

rjou

rnal

s.an

nual

revi

ews.

org

by U

NIW

ER

SYT

ET

WR

OC

LA

WSK

I on

04/

18/0

8. F

or p

erso

nal u

se o

nly.


and FUT8 from family GT23 (63), and theGlcNAc transferase essential for bacterial cellwall synthesis known as MurG of familyGT28 (64, 65). These enzymes all appear touse metal ion-independent methods for sta-bilizing the departure of nucleoside diphos-phate leaving groups analogous to that de-scribed above. In the cases of WaaC and FucT,structural and kinetic analyses of site-directedmutants support the roles of the side chain car-boxylates of Asp13 and Glu95, respectively,as base catalysts. In the cases of FUT8 andMurG, the identity of the base remains lessclear.

A metal ion-independent sialyltransferase,PmST1 from family GT80, has been reportedto possess four distinct enzymatic activitieswith differing pH optima consisting of anα-2,3-sialyltransferase activity with an opti-mal pH of 7.5–9.0, an α-2,6-sialyltransferaseactivity with an optimal pH of 4.5–6.0, anα-2,3-sialidase activity with an optimal pHof 5.0–5.5, and an α-2,3-trans-sialidase ac-tivity with an optimal pH of 5.5–6.5 (66). Itshould be noted that the adventitious pres-ence of CMP could be the cause of the ob-served sialidase and trans-sialidase activities,as this enzyme, like most other glycosyltrans-ferases, presumably catalyzes the hydrolysis ofits donor sugar substrate. At the high con-centrations of enzyme used in determiningthese activities, back reaction in the pres-ence of CMP would lead to the formation ofCMP NeuAc in situ, which could then be hy-drolyzed (leading to the observed sialidase ac-tivity) or used as a substrate in the catalyzedtransfer of sialic acid to an alternative accep-tor (leading to the observed trans-sialidase ac-tivity). In fact, it has recently been elegantlyshown that glycosyltransferase reactions canbe run in reverse to generate desired rare nu-cleoside diphosphosugars in situ that can thenbe used to generate a range of glycosylated an-tibiotic products (67). The 3-D X-ray crystalstructure of PmST1 has been solved in apoand product complex form with bound CMPunder conditions (pH 7.5) that favor the α-2,3-sialyltransferase activity (68). Subsequent

cocrystallization with CMP 3F NeuAc, fol-lowed by soaking with a lactose acceptor, ledto the production of a ternary complex struc-ture (69).

Mechanism of RetainingGlycosidases

It is important to discuss the mechanism ofretaining glycosyltransferases in the contextof the vast body of work on retaining glyco-side hydrolases. With the notable exceptionof members from families GH4 and GH109,which use a recently described eliminationmechanism involving transient remote oxida-tion (70, 71) and retaining β-hexosaminidasesthat use an intramolecular nucleophile (re-viewed in Reference 72), the typical mech-anism of retaining glycosidases is that of adouble-displacement reaction involving a co-valently bound glycosyl-enzyme intermediatespecies (3). This mechanism was first put for-ward by Koshland (73), who realized that,because inversion of configuration is a fun-damental and universal property of bimolec-ular displacement at saturated carbon centers,glycosidases that gave sugar products with thesame anomeric configuration as the substratemust catalyze the reaction using two distinctnucleophilic displacement steps involving anenzymatic catalytic nucleophile. An alterna-tive SN1-like mechanism, involving the for-mation of a discrete enzyme-stabilized ox-ocarbenium ion intermediate species that isshielded on one face by the enzyme, therebypreventing nucleophilic attack from the op-posite face of the reaction center and lead-ing to complete retention of anomeric con-figuration in the product, was subsequentlyproposed for the retaining glycosidase henegg white lysozyme by Phillips (74, 75). Thefree energy of the intermediate, and there-fore the associated transition states, for theKoshland mechanism would be lower than forthose associated with the Phillips mechanism.The glycosidic linkage between two sugars isextremely stable. For example, the half-livesfor the spontaneous hydrolysis of starch and


Ann

u. R

ev. B

ioch

em. 2

008.

77. D

ownl

oade

d fr

om a

rjou

rnal

s.an

nual

revi

ews.

org

by U

NIW

ER

SYT

ET

WR

OC

LA

WSK

I on

04/

18/0

8. F

or p

erso

nal u

se o

nly.


cellulose at room temperature and neutral pHare in the range of five million years (76). Assuch, the fact that glycosidases are observed tocatalyze the hydrolysis of these materials withrate constants of up to 1000 s−1 would leadone to believe that this class of enzyme hasevolved a mechanism that would involve anintermediate species with the lowest free en-ergy possible without making turnover of thisspecies too slow. The attendant decreases inthe activation barrier allow these enzymes toachieve the formidable task of glycoside cleav-age with such proficiency. With some notableexceptions (77), this logic and a large arrayof mechanistic data (for example, References78–83) has led most in the field to accept theKoshland mechanism as the mechanism uti-lized by virtually all retaining glycosidases.

Mechanism of RetainingGlycosyltransferases

Again by direct comparison to retaining gly-cosidases, the mechanism of retaining glyco-syltransferases has been proposed to be that ofa double-displacement mechanism involvinga covalently bound glycosyl-enzyme interme-diate (Figure 4b), demanding the existenceof an appropriately positioned nucleophilewithin the active site (2). A divalent cationor suitably positioned positively charged sidechains or helix dipoles would presumably playthe role of the Lewis acid as was describedabove for the inverting glycosyltransferases.The leaving diphosphate group itself probablyplays the role of a base catalyst activating theincoming acceptor hydroxyl group for nucle-ophilic attack. The role of substrate/productphosphates acting as base catalysts has a richhistory and, as an example, has recently beenpostulated in the mechanisms of the farnesyldiphosphate (FPP) synthases from Escherichiacoli (84) and Trypanosoma cruzi (85). In contrastto glycoside hydrolases, however, there is lit-tle work that supports a double-displacementmechanism as a canonical glycosyltransferasemechanism.

Challenges in Studying theMechanisms of RetainingGlycosyltransferases

Mechanistic characterization of this class ofenzymes has proven to be a challenging task.The conclusive identification of a catalytic nu-cleophile and observation of a true kineticallyand catalytically competent covalent interme-diate has yet to be reported for any retainingtransferase despite years of exhaustive stud-ies using techniques that have been success-fully applied to the characterization of retain-ing glycosidases. This may well be interpretedas evidence against the double-displacementmechanism, but it could also be the result ofthe inapplicability of these techniques for thestudy of transferases owing to inherent dif-ferences in the nature of the substrates be-ing studied. For example, the most successfulapproach used for the characterization of re-taining glycosidases has involved the use offluorinated substrate analogues (86). The in-troduction of an electronegative fluorine ateither the C2 or C5 position of a pyranosering inductively destabilizes the oxocarbe-nium ion-like transition states through whichboth steps of the double-displacement reac-tion proceed and in some cases also removeskey hydrogen-bonding interactions, resultingin a significant decrease in the rate of theoverall reaction. By introducing a good leav-ing group (e.g., dinitrophenol or fluoride), thefirst step is “rescued,” resulting in the accu-mulation of the intermediate species with asignificant lifetime that allows mass spectro-metric and X-ray crystallographic characteri-zation. Alternatively, or used in combinationwith the fluoro-sugar approach, removal ofthe acid/base catalyst of a retaining glycosi-dase by mutagenesis also leads to a decreasein the rates of both steps of the reaction, andagain, by using a substrate with a highly ac-tivated leaving group, the glycosylation stepcan be rescued, leading to the accumulationof the glycosyl-enzyme intermediate.

However, because of the strict glycosyl-transferase requirement for their nucleoside


Ann

u. R

ev. B

ioch

em. 2

008.

77. D

ownl

oade

d fr

om a

rjou

rnal

s.an

nual

revi

ews.

org

by U

NIW

ER

SYT

ET

WR

OC

LA

WSK

I on

04/

18/0

8. F

or p

erso

nal u

se o

nly.


b Glycogenin

d ppGalNAcT10

f MGS

D190

UDP 2F Gal

4'-Deoxy lactose

Y151

D130

Y278 E317

H280

D225

R202

Q346Y347

E345

R221

GalNAcUDP

D237

H370

Y372

H239

R373

D316

S318

UDP 2F Gal

D227

Q385

N384

R273

S469

L163(side chain

omitted)

E166 D100

K215

H217 GDP man

K76

D192

T193

D270D286

Glc

UDP

E515

Q189

D188

R86 G165

Q164

D125 N133

H212

S1346.1

D163

K86

UDP Glc

D104

K218

8.9

3.5

4.5

3.5

4.3

3.4

6.1

(pdb 1ll2)(GT8)

a LgtC

c 3GalT

e ToxB

(pdb 1ga8)(GT8)

(pdb 2vfz)(GT6)

(pdb 2d7r)(GT27)

(pdb 2bvl)(GT44) (pdb 2bo8)

(GT78)


Ann

u. R

ev. B

ioch

em. 2

008.

77. D

ownl

oade

d fr

om a

rjou

rnal

s.an

nual

revi

ews.

org

by U

NIW

ER

SYT

ET

WR

OC

LA

WSK

I on

04/

18/0

8. F

or p

erso

nal u

se o

nly.


diphosphate (NDP) leaving group, the rel-ative leaving group ability cannot so easilybe manipulated, thus the relative rates of theglycosylation and deglycosylation steps can-not be altered. In order to prevent hydrol-ysis of metabolically expensive (substituted)phosphodonor sugar substrates, it would beexpected that glycosyltransferases would haveevolved a mechanism in which the step involv-ing cleavage of the bond between the sugarand its activated leaving group would be ratelimiting thereby preventing the accumulationof an (observable) intermediate species, whichcould be easily hydrolyzed. In addition, be-cause the leaving groups are themselves be-lieved to play the role of base catalyst, thedeglycosylation step cannot be slowed downfor this class of enzyme by simple mutagene-sis of the base catalyst. This inability to alterthe relative rates of glycosylation versus deg-lycosylation has rendered the fluoro sugar ap-proach ineffective in the trapping of interme-diates on retaining transferases (87). Possiblemechanistic proposals are considered, below,in light of 3-D structural characterization ofretaining glycosyltransferases.

Retaining GT-A Glycosyltransferases

Structural and mechanistic studies indicategenerally utilized strategies among membersof retaining GT-A fold glycosyltransferasesfacilitating leaving group departure and acti-vating the nucleophilic group of the incom-ing acceptor. However, the observed struc-tures have not revealed a conserved structuralarchitecture on the β-face of the donor sugarsubstrate in the region that would be expectedto be occupied by the obligate enzymatic nu-cleophile of a proposed double-displacementmechanism. This implies either a different re-

UDP 2F Gal:uridine5′-diphospho-(2-deoxy-2-fluoro)-α-D-galactopyranose

action mechanism or the need for substantialand different conformational changes for eachenzyme.

Using uridine 5′-diphospho-(2-deoxy-2-fluoro)-α-D-galactopyranose (UDP 2F Gal)and 4′-deoxy lactose as the nonreactive donorand acceptor substrate analogues, respec-tively, a well-resolved ternary complex struc-ture was obtained for the family GT8 galacto-syltransferase LgtC from Neisseria meningitidis(88) (Figure 6a). As was seen with the invert-ing GT-A enzymes, this structure revealed thepresence of a Mn2+ cation, coordinated withinthe active site by the carboxylate side chainsof a DXD motif and by the donor substratediphosphate leaving group, which presumablyacts as a Lewis acid to facilitate leaving groupdeparture. In the region that would be occu-pied by the reactive axial 4′-hydroxyl of the ac-ceptor substrate analogue, the only functionalgroup that is suitably positioned to activatethe incoming nucleophile is the leaving groupoxygen of the donor substrate. This suggeststhat the departing diphosphate moiety acts asthe base catalyst. Surprisingly, the functionalgroup most suitably positioned within the ac-tive site on the β-face of the donor substrate toplay the role of the catalytic nucleophile is thatof the side chain amide of Gln189. The amidecarbonyl oxygen of this residue is perfectly po-sitioned 3.5 A away and has an ideal trajec-tory for nucleophilic attack on the anomericreaction center. However, the correspondingalanine mutant retained 3% activity, inconsis-tent with an essential role for this residue asthe catalytic nucleophile.

The possibility of a double-displacementmechanism involving nonenzymatic nucle-ophiles has also been investigated with LgtCbecause one of the lactose hydroxyls wasintriguingly positioned for such a role. In

←−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−Figure 6Comparison of the active sites of several retaining GT-A fold glycosyltransferases. Descriptions ofconserved structural features, and a lack thereof, are provided in the text. Mn2+ cations, which play therole of a Lewis acid catalyst that facilitates leaving group departure, are shown as purple spheres.Distances are indicated in angstroms.


Ann

u. R

ev. B

ioch

em. 2

008.

77. D

ownl

oade

d fr

om a

rjou

rnal

s.an

nual

revi

ews.

org

by U

NIW

ER

SYT

ET

WR

OC

LA

WSK

I on

04/

18/0

8. F

or p

erso

nal u

se o

nly.


order to achieve overall net retention of theanomeric configuration, a hydroxyl groupfrom one of the substrates could act as a nucle-ophile that displaces the UDP leaving groupin an initial step. Subsequent attack on theresulting β-linked galactosyl intermediate bythe 4′-hydroxyl of the lactose acceptor fromthe α-face would yield the resulting Gal-(α-1,4)-Gal linkage. A mechanism involving thelactose acceptor in this way would have the ad-vantage of minimizing unwanted donor sub-strate hydrolysis, as it would obligate ternarycomplex formation prior to the production ofa reactive intermediate species. Limiting theplausibility of such a mechanism is the factthat the resulting intermediate would be oneof an inherently unreactive glycoside species.This mechanism has been discounted as a re-sult of the finding that the chemically syn-thesized putative intermediates are not turnedover by the enzyme (88).

To investigate the degree of nucleophiliccharacter contributed by Gln189 during catal-ysis, a Q189E variant of LgtC was created.This substitution would introduce a better nu-cleophile and a worse leaving group for theputative covalent glycosyl-enzyme interme-diate (pKa of −0.5 to −1.0 for protonationof the oxygen of an amide versus a pKa of∼4.0 for that of a carboxylate), which maylead to a rate-limiting deglycosylation stepand the potential accumulation of the inter-mediate species. The results of this work in-deed led to the first direct observation of a cat-alytically relevant covalent glycosyl-enzymeintermediate for a retaining glycosyltrans-ferase (89). However, the site of labeling wasfound to be a residue (Asp190) that is rel-atively remote from the anomeric reactioncenter in the ground state crystal structure.Support for a critical catalytic role for thisresidue was provided by kinetic analysis of theD190N mutant, which revealed a 3000-folddecrease in the observed turnover rate com-pared to the wild-type enzyme. However, ifthis residue were to act as a catalytic nucle-ophile in a double-displacement mechanism,a significant conformational change from that

of the ternary complex crystal structure wouldhave to occur during the course of catalysis.Furthermore, the lack of sequence or struc-tural conservation implies that not only is aconformational change required but also thatthis change would have to be different forrelated enzymes, even those with substantialsimilarities to LgtC.

Subsequently, the 3-D X-ray crystal struc-ture of rabbit muscle glycogenin, another en-zyme of the GT8 family, with an intact UDPGlc donor substrate bound within the activesite was reported (90). This enzyme catalyzesself-glucosylation of a tyrosine residue in an-other monomer of the enzyme; plus succes-sive glycosylations of the glucosyl tyrosineformed in a process that is the initial step ofglycogen biosynthesis. As was the case withLgtC, a Mn2+ cation was observed withinthe active site of glycogenin coordinated bythe side chain carboxylates of a DXD mo-tif positioned to act as a Lewis acid thatactivates the departing diphosphate leavinggroup (Figure 6b). However, unlike LgtC,the side chain most suitably positioned onthe β-face of the donor substrate to act asthe catalytic nucleophile is that of the car-boxylate of Asp163, albeit at a distance of6.1 A from the reaction center (Figure 6b).This conserved residue, which correspondsto Asp188 of LgtC, is in close proximity tothe positively charged side chain of Lys86,and this residue pairing constitutes the onlyconserved structural motif that can be iden-tified on the β-face of the donor substrate-binding sites of retaining glycosyltransferases.In LgtC, the side chain carboxylate of Asp188is within hydrogen-bonding distance of theC6 and C4 hydroxyls of the galactose moi-ety of the donor substrate and the positivelycharged side chain of Arg86 (Figure 6a). Inthe glycogenin structure, the side chain amideof Gln164, which corresponds to Gln189 inLgtC, is within hydrogen-bonding distanceof the C4 hydroxyl of the glucose donor sub-strate (Figure 6b). These differences in rel-ative positioning indicate that either differ-ent donor sugar-binding modes exist among


Ann

u. R

ev. B

ioch

em. 2

008.

77. D

ownl

oade

d fr

om a

rjou

rnal

s.an

nual

revi

ews.

org

by U

NIW

ER

SYT

ET

WR

OC

LA

WSK

I on

04/

18/0

8. F

or p

erso

nal u

se o

nly.


enzymes of the same family and/or that signif-icantly different ground state structures exist,indicating a high degree of structural plastic-ity for this class of enzyme. As a note of cau-tion, in the described glycogenin structure,the glucose (from UDP Glc) is proposed tobind in a totally different orientation with re-spect to the UDP sugar moiety compared tothat seen on all other GT-A fold retaining gly-cosyltransferases. Furthermore, this proposalis based upon substantially weaker electrondensity for the glucose than for the surround-ing protein environment and UDP.

Bovine α3GalT from family GT6 is an-other retaining galactosyltransferase possess-ing a GT-A fold that has been the subjectof significant structural and mechanistic in-vestigations. Based on an initial 3-D X-raycrystal structure, Glu317 (structurally equiv-alent to Gln 189 of LgtC) was proposed toact as the catalytic nucleophile in a double-displacement mechanism (91). In fact, in thisinitial report, the authors claimed to observea covalently bound intermediate species. Thisclaim was made on the basis of a poor elec-tron density map. This, in combination witha lack of order in the active-site region and thepresence of two different crystal forms for thestructure containing the proposed covalentintermediate, has led to a general rejection ofthis interpretation. Later, a detailed study ofthe binding affinities of wild-type and E317Qenzymes for donor and acceptor substratesled to the proposition that this residue ismore likely required for proper acceptor sub-strate orientation (92). However, in contrastto the mutagenesis results with the equivalentGln189 residue in LgtC, this study revealedthat mutation of Glu317 leads to a 2400-folddecrease in the observed turnover rate, indi-cating the critical catalytic importance of thisresidue (92). A structure with the nonreac-tive donor substrate analogue UDP 2F Galbound within the active site was recently ob-tained and revealed the suitable glycosidase-like positioning of the side chain carboxylateof Glu317 poised to play the role of a cat-alytic nucleophile (93) (Figure 6c). In addi-

tion, it has been shown that the activity of theE317A mutant of α3GalT can be rescued bythe addition of small exogenous nucleophilessuch as azide, and indeed, a β-glucosyl azidederivative was isolated and characterized (94).Chemical rescue has been successfully em-ployed to identify the acid/base and nucle-ophile catalytic residues of retaining glycosi-dases. As such, the chemical rescue of Glu317is suggestive that this side chain is suitably po-sitioned to play such a catalytic role in themechanism of α3GalT. However, it must bekept in mind that, although chemical rescuedoes indicate that a side chain is suitably posi-tioned to play a catalytic role, it does not provethat it plays such a role in the natural enzymemechanism! It simply indicates the likely po-sitioning of that residue with respect to the re-action center. It is not perhaps surprising thatwhen reactions of a mutant enzyme, havingan introduced space into which a good nucle-ophile can bind within an enzyme active sitein the vicinity of a reactive electrophilic cen-ter, are run in the presence of high concentra-tions of a small nucleophile that the enzyme isfound to act as a scaffold that catalyzes an un-natural reaction simply by proximity effects.Chemical rescue results alone cannot be usedto deduce the natural catalytic mechanism ofan enzyme.

The conserved structural motif, involv-ing the side chain carboxylate of Asp316within hydrogen-bonding distance of donorsugar hydroxyls and the positively chargedside chain of residue Arg202, is also presentin α3GalT (Figure 6c). The α3GalT struc-tures also reveal the presence of a bound di-valent cation within the active site suitablypositioned to play the role of a Lewis acidthat facilitates leaving group departure. In ad-dition, a product complex with bound UDPcontained a second Mn2+ bound within theactive site in a way that suggests a role in fa-cilitating product release (93).

The glycosyltransferases responsiblefor addition of the specificity-determiningGal/GalNAc moieties to the cell surfaceglycolipids and glycoproteins that constitute


Ann

u. R

ev. B

ioch

em. 2

008.

77. D

ownl

oade

d fr

om a

rjou

rnal

s.an

nual

revi

ews.

org

by U

NIW

ER

SYT

ET

WR

OC

LA

WSK

I on

04/

18/0

8. F

or p

erso

nal u

se o

nly.


the ABO(H) blood groups are another groupof family GT6 enzymes that have beensubjected to intensive structural scrutiny (95,96). Glycosyltransferase A (GTA) specificallyuses UDP GalNAc as a donor substrate tomodify the 3-OH of the terminal galactoseresidue of the H antigen [terminal Fuc-α(1,2)-Gal-β-R acceptors], thereby creatingthe A blood group antigen. Conversely,glycosyltransferase B (GTB) specifically usesUDP Gal to modify the same hydroxyl ofH antigen to generate the B blood groupantigen. Remarkably, these two enzymesdiffer from each other by only 4 out of354 amino acids. The 3-D X-ray crystalstructures of these two enzymes with bothUDP and H antigen bound revealed thatdistinction between the two donor substratesis achieved by the substitutions of residuesLeu266 and Gly268 in GTA with the morebulky side chains of Met266 and Ala268 inGTB (97). Similarly, a crystal structure ofthe inactive mutant of this enzyme, expressedby individuals of the O blood group whereinterminal H antigens are not modified, re-vealed how a single mutation that changesGly/Ala268 with the bulky side chain of Argblocks donor substrate binding and resultsin the ablation of catalytic activity (98). Therole of intramolecular hydrogen bondingwithin acceptors (99) and the role of the tworesidues (Gly/Ser235 and Leu/Met266) (100)in the assembly of Type I and II H antigensby GTA and GTB have also been revealed byX-ray crystallographic analysis. Finally, therole of various missense mutations leadingto the production of enzymes of weak dualspecificity has been investigated (101, 102).

The GT27 family contains a series ofUDP GalNAc:polypeptide α-N-acetyl-galactosaminyltransferases (ppGalNAcTs)that initiate the formation of mucin-typeO-linked glycans by catalyzing the transferof GalNAc to serine or threonine residueson protein surfaces, thereby generating theTn antigen (GalNAc-α-O-Ser/Thr). Theseenzymes are particularly interesting in thatthey possess C-terminal lectin domains. The

majority of ppGalNAcTs, termed peptidetransferases, are able to catalyze transferto both unmodified peptides as well as toglycopeptides. Others, termed glycopeptidetransferases are only able to transfer topeptides that have been modified by an initialGalNAc residue (103, 104). The 3-D X-raycrystal structures of murine ppGalNAcT-1isozyme with only Mn2+-bound (105), humanppGalNAcT-10 isozyme in complex with ei-ther a hydrolyzed donor or a GalNAc-serineacceptor (106) and a human ppGalNAcT-2isozyme with UDP and a bound acceptorpeptide (107) have been reported. Thesestructures indicate that activation of theUDP leaving group is achieved in the usualfashion by the presence of a coordinatedMn2+ cation within the active site poised toact as a Lewis acid (Figure 6d ). In addition,the conserved motif, comprising a side chaincarboxylate (Glu345 in ppGalNAcT-10and Glu334 in ppGalNAcT-2) positionedwithin hydrogen-bonding distance of donorsubstrate hydroxyls and a positively chargedside chain (Arg221 in ppGalNAcT-10 andArg208 in ppGalNAcT-2), was again ob-served (Figure 6d ). The side chain amidesof Gln346 of ppGalNAcT-10 and Asn335 ofppGalNAcT-2 are located in a structurallyanalogous position to that of Gln189 in LgtCand are therefore the most likely candidatesto act as the catalytic nucleophiles on thebasis of available structural informationonly (Figure 6d ). These structural studieshave also revealed how dynamic associationbetween the lectin and catalytic domains,facilitated by a flexible tether, permits theformation of a range of binding modes forvarious glycopeptide acceptor substratesthereby facilitating the production of thecharacteristic high-density glycosylationpatterns associated with mucins.

The catalytic domain of Clostridium difficiletoxin B (ToxB) is a retaining family GT44 glu-cosyltransferase possessing a GT-A fold thatmodifies host Rho proteins by glycosylatingkey threonine residues (108). A 3-D X-raycrystal structure of the catalytic domain with


Ann

u. R

ev. B

ioch

em. 2

008.

77. D

ownl

oade

d fr

om a

rjou

rnal

s.an

nual

revi

ews.

org

by U

NIW

ER

SYT

ET

WR

OC

LA

WSK

I on

04/

18/0

8. F

or p

erso

nal u

se o

nly.


bound Mn2+, UDP, and glucose has been re-ported (109). This structure revealed the con-served GT-A retaining architectural featuresof a bound Mn2+ that acts as a Lewis acidactivator as well as a side chain carboxylate(Asp270) within hydrogen-bonding distanceof the donor sugar hydroxyls and a positivelycharged side chain (Arg273) (Figure 6e). Theside chain amide of Asn384 is the most suit-ably positioned candidate for a putative cat-alytic nucleophile.

Mannosylglycerate synthase (MGS) fromRhodothermus marinus is a family GT78 man-nosyltransferase that is responsible for thesynthesis of the stress protectant 2-O-α-D-mannosylglycerate. An X-ray crystal struc-ture of MGS with Mn2+ and intact GDPMan bound within the active site also revealedthe presence of a coordinated divalent cationLewis acid activator and a side chain carboxy-late (Asp192) within hydrogen-bonding dis-tance of donor sugar hydroxyls and a posi-tively charged side chain (Lys72) (Figure 6f )(27). The main chain amide of Leu163 is theonly functional group suitably positioned onthe β-face of the anomeric reaction center ofthe mannosyl donor substrate to act as a cat-alytic nucleophile (Figure 6f ).

Three-dimensional X-ray crystal struc-tures have been reported for the retainingenzymes Kre2 (110), a yeast mannosyltrans-ferase from family GT15, and EXTL2 (111),a mouse N-acetylhexosaminyltransferase in-volved in heparan biosynthesis from familyGT64. Both of these enzymes were shownto possess a GT-A fold; however, the absenceof bound donor sugar substrates limits theutility of these structures in identifying can-didate catalytic nucleophiles. Both reportedstructures had bound divalent cations coor-dinating the nucleoside diphosphate product.A ternary complex structure of EXTL2 withbound UDP and an acceptor substrate sup-ports a role for the phosphate leaving groupin acting as the base catalyst that activates theincoming nucleophile.

A comparison of representative GT-A foldretaining glycosyltransferases from various

families illuminates conserved structural fea-tures within some regions of the active site anda lack thereof in other regions for this class ofenzyme. The mode of leaving group activa-tion is similar to that used by the majority ofinverting GT-A enzymes. A divalent cation,coordinated by the side chains of a conservedDXD motif, acts as a Lewis acid that facilitatesleaving group departure. In contrast to theconserved features in the region surroundingthe leaving group and incoming acceptor (theα-face of the donor substrate), there appearto be very few conserved architectural fea-tures within the active-site region that accom-modates the β-face of the donor. The onlystrictly conserved feature observed on thisface of the reaction center is a side chain car-boxylate (from an Asp/Glu residue typicallysituated on helix 6) that is within hydrogen-bonding distance of the donor sugar hydroxylsand a positively charged side chain. The sidechain of this Arg/Lys residue is in turn hy-drogen bonded to a side chain carboxylate,derived from a residue of the DXD motif,which is in turn coordinated to the bounddivalent cation. Interestingly, in the case ofglycogenin, the side chain from this β-faceresidue (Asp163) is the most suitably situatedgroup to act as the catalytic nucleophile ina double-displacement mechanism, indicatinghow subtle differences in the mode of donorsugar binding can influence our interpreta-tion of a mechanism when based solely onstructural information. In most cases, a sidechain or main chain amide is most suitablypositioned in an appropriate trajectory and ata reasonable distance from the anomeric re-action center to play the role of the catalyticnucleophile in a putative double-displacementmechanism. An exception to this observationis found with family GT6 enzymes that havea side chain carboxylate at this position.

This apparent lack of conserved active-site architecture among retaining GT-A gly-cosyltransferases, all of which appear to havediverged from a common ancestor, is instark contrast to what has been observedfor the analogous retaining glycosidases. In


Ann

u. R

ev. B

ioch

em. 2

008.

77. D

ownl

oade

d fr

om a

rjou

rnal

s.an

nual

revi

ews.

org

by U

NIW

ER

SYT

ET

WR

OC

LA

WSK

I on

04/

18/0

8. F

or p

erso

nal u

se o

nly.


those cases, despite having a multitude ofobserved 3-D folds, highly conserved pairsof carboxylates are situated within the ac-tive site, with one clearly positioned to playthe role of catalytic nucleophile. This wouldindicate a convergence in mechanisms amongretaining glycosidases. If all retaining glyco-syltransferases utilize the analogous double-displacement mechanism, it would seem likelythat during the course of divergent evolutionthe most stringently conserved active-site fea-ture would have been the relative positioningof the most important component of the cat-alytic machinery. On the basis of the availablecrystal structures, this feature appears to beabsent. This observation may very well sug-gest that retaining glycosyltransferases utilizea mechanism that differs from that of retain-ing glycosidases, as discussed below.

Retaining GT-B Glycosyltransferases

As is the case for the inverting glycosyltrans-ferases possessing a GT-B fold, retaining en-zymes with the GT-B fold also use a metal ion-independent method for facilitating leavinggroup departure. Crystal structures with sub-strates or inhibitors bound within the donorsugar site reveal the conserved presence of twoactive-site functional groups on the β-face ofthe donor substrate, as noted below.

Members of family GT35 catalyze thephosphorolysis of α(1,4)-linked glucans lead-ing to the production of α-D-glucose-1-phosphate (Glc1P), which is in turn isomer-ized to glucose-6-phosphate and fed into theglycolytic pathway. The enzymes of this fam-ily play an essential role in energy storage/mobilization in organisms ranging from bac-teria to mammals and have a noteworthyhistory of investigation, with a Nobel Prizeawarded to Carl and Gerty Cori in 1947for their studies of rabbit muscle glycogenphosphorylase (rmGP) (112) and another toEdmond Fischer and Edwin Krebs in 1992for the first discovery of protein phospho-rylation as a control mechanism on eukary-otic members of this family (113, 114). By

contrast, the activities of the bacterial en-zymes are controlled at the level of expression(115).

The most intensively studied member offamily GT35 is that of rmGP, with over 30structural investigations having been reportedto date (for examples, see References 116 and117). However, despite a long history of ex-haustive and laborious research with rmGP, adetailed understanding of the catalytic mech-anism remains elusive. Crystal structures arealso available for bacterial maltodextrin phos-phorylase (MalP) (118–121), yeast glycogenphosphorylase (122), human muscle glycogenphosphorylase (123), and human liver glyco-gen phosphorylase (124–127).

Members of this family are unique amongthe glycosyltransferases in that they contain apyridoxal phosphate group covalently boundwithin their active site via a Schiff base to alysine residue. The current view is that pyri-doxal phosphate acts as a surrogate for the nu-cleoside monophosphate portion of a nucle-oside diphosphate lost during the evolutionfrom an NDP sugar-dependent GT-B glyco-syltransferase to a GT-B glycosyltransferase-like phosphorylase (128–131).

A ternary complex 3-D X-ray crystal struc-ture of rmGP, with the putative transitionstate analogue nojirimycin-tetrazole and withphosphate also bound within the active site,revealed the main chain amide of His377 tobe the functional group most suitably po-sitioned to play the role of a catalytic nu-cleophile (117), should one invoke such amechanism. Soaking (Glc)4 or (Glc)5 malto-oligosaccharides into crystals of E. coli MalPwith bound GlcP resulted in the productionof ternary complex structures with both trans-fer product and phosphate bound within theactive site (120). Like rmGP, the main chainamide of His345 (corresponding to His377 inrmGP) was found to be the most suitably po-sitioned candidate to act as the catalytic nu-cleophile (Figure 7a). The only other suitablefunctional group positioned on the β-face ofthe donor substrate to act as a catalytic nucle-ophile is the side chain carboxylate of Asp307


Ann

u. R

ev. B

ioch

em. 2

008.

77. D

ownl

oade

d fr

om a

rjou

rnal

s.an

nual

revi

ews.

org

by U

NIW

ER

SYT

ET

WR

OC

LA

WSK

I on

04/

18/0

8. F

or p

erso

nal u

se o

nly.


H345 D307

GlcGlc

K539

R534PLP

D130

H154

UDP Glc

H116

D115

H140

H114

UDP Glc

R204

K209

E306D361

K267

R262

R208

E281

D100

A99

D19

UDP 2F Glc

K209

3.3

2.8

2.9

5.5 2.83.0

2.6

2.7

3.1

2.8

5.64.9

3.0

2.3

5.8

4.6

3.2

2.7

3.4

2.8

6.34.1

2.8

2.8 2.9

2.7

a MalP

c OtsA

(pdb 1l6i)(GT35)

(pdb 1uqu)(GT20)

b WaaG

d AGT

(pdb 2iw1)(GT4)

(pdb 1y6f)(GT72)

Figure 7Comparison of the active sites of several retaining GT-B fold glycosyltransferases. Distances areindicated in angstroms.

(Figure 7a). However, this carboxylate issituated within hydrogen-bonding distanceof the C2- and C3-hydroxyls of the accep-tor glucose residue and therefore most likelyserves to recognize and orient the incomingacceptor.

From the MalP structure, it is apparentthat formation of a hydrogen bond between

the side chain imidazole of His 345 and theC6-hydroxyl of what would have been the glu-cose donor sugar results in the closure of aflexible loop region that makes up a signifi-cant portion of the maltose acceptor recogni-tion site (Figure 7a). Analogous to the elec-trostatic strategy of inverting GT-B enzymes,the bound phosphate product is located within


Ann

u. R

ev. B

ioch

em. 2

008.

77. D

ownl

oade

d fr

om a

rjou

rnal

s.an

nual

revi

ews.

org

by U

NIW

ER

SYT

ET

WR

OC

LA

WSK

I on

04/

18/0

8. F

or p

erso

nal u

se o

nly.


hydrogen-bonding distances of the positivelycharged MalP side chains of Arg534 andLys539, indicating a role for these residuesin stabilizing the departing phosphate leavinggroup (Figure 7a).

Three-dimensional X-ray crystal struc-tures with bound intact donor substrates/analogues are also available for retainingGT-B fold glycosyltransferases from fami-lies GT4, GT20, and GT72. The structureof E. coli α-(1,3) glucosyltransferase WaaGfrom family GT4, involved in lipopolysac-charide biosynthesis, was solved with boundnonreactive donor sugar substrate analogueuridine 5′-diphospho-(2-deoxy-2-fluoro)-α-D-glucopyranose (UDP 2F Glc) (Figure7b) (132). Such 2-deoxy-2-fluorosugar ana-logues of NDP-sugars have proved valuabledonor substrate analogues that do not un-dergo reaction at the glycosyltransferase ac-tive site because the fluorine inductively desta-bilizes transition states for glycosyl trans-fer, effectively prohibiting reaction. Only inthe case of GT80 sialyl transferases has anyturnover been observed with such fluoro-sugar analogues (69). Similarly, structuresof E. coli OtsA from family GT20, respon-sible for the biosynthesis of the very in-teresting stress response molecule α,α-1,1trehalose-6-phosphate, were obtained witheither UDP 2F Glc or UDP Glc (Figure 7c)bound (133). Finally, ternary complex struc-tures of AGT from family GT72, the re-taining DNA-modifying glucosyltransferasecounterpart from the T4 bacteriophage ofthe inverting BGT, were obtained with UDPGlc and various DNA acceptor substratesbound (134) (Figure 7d ). Like BGT, AGTwas shown to use a “base-flipping” mechanismto activate DNA acceptors. Crystal structuresof bacterial (135) and archael (136) glyco-gen phosphorylases from family GT5 are alsoavailable. However, the absence of the sugarmoiety of the donor substrate limits theirutility in identifying candidate nucleophilicresidues.

A comparison of these structures, alongwith those of the rmGP and MalP ternary

complexes, facilitates an initial identificationof several conserved active-site structural fea-tures among GT-B fold glycosyltranferases.On the β-face of the bound donor sugarsubstrate, a main chain amide is most suit-ably positioned, at an appropriate distanceand trajectory, to act as the catalytic nucle-ophile in a double-displacement mechanism.The conserved presence of hydrogen bondingpartners, donated from main chain carbonylor side chain amide groups, in proximity tothe main chain NH group could help facili-tate this catalytic role. The only other candi-date for the role of nucleophile is that of theside chain carboxylate of a structurally con-served aspartate (Asp307, Asp100, Asp130,Asp115 in MalP, WaaG, OtsA, and AGT,respectively, Figure 7) situated ∼6 A fromthe anomeric reaction center. This residueis more likely involved in acceptor substraterecognition/orientation, and the plausibilityof a nucleophilic role for this residue isseverely limited by mutagenesis investiga-tions with AGT in which it was found thatthe D115A mutant displayed ∼10% resid-ual transferase activity (134). Another con-served feature on the β-face of the donoris the hydrogen bond formed between thedonor sugar C6-hydroxyl and, with the ex-ception of WaaG where Asp19 is the bond-ing partner, the side chain imidazole of theHis residue whose main chain amide is posi-tioned to act as a nucleophile (His354/377,His154, and His114 in MalP/rmGP, OtsA,and AGT, respectively, Figure 7). As men-tioned above, the formation of this interac-tion is believed to cause the closure of a flex-ible loop leading to the complete formationof the acceptor recognition site. On the α-face of the donor substrate, a structurally con-served Arg/Lys pair of positively charged sidechains is present within hydrogen-bondingdistance of the phosphate leaving group.These are poised to replace the role of a di-valent cation in GT-A fold enzymes, electro-statically stabilizing the leaving group depar-ture (Figure 7). The ternary complex struc-ture of MalP with bound products suggests


Ann

u. R

ev. B

ioch

em. 2

008.

77. D

ownl

oade

d fr

om a

rjou

rnal

s.an

nual

revi

ews.

org

by U

NIW

ER

SYT

ET

WR

OC

LA

WSK

I on

04/

18/0

8. F

or p

erso

nal u

se o

nly.


Figure 8(a) The SNi mechanism for the decomposition of alkyl chlorosulfites. (b) A comparison of theintermediates involved in the SNi mechanism for the decomposition of alkyl chlorosulfites and in theproposed mechanisms for the solvolysis of glycosyl fluorides and of retaining glycosyltransferases.

that the substrate/product phosphate group ismost suitably positioned to play the role ofthe base catalyst that activates the functionalgroup of the incoming acceptor (Figure 7a).

An Alternative SNi-Like Mechanism

Albeit rather rare, reactions involving a sin-gle nucleophilic displacement step can leadto the formation of a product in which thestereochemistry of the reaction center is com-pletely retained. This is a special case of theSN1 mechanism that involves the formationof discrete ion pair intermediates, with life-times longer than a bond vibration, which cancollapse to give back the starting material or aproduct (Figure 8a). This unique form of SN1reaction, termed SNi wherein the “i” indicatesinternal return, was first described to accountfor the observed products and the nature ofthe reaction involved with the decompositionof alkyl chlorosulfites (137–139). The uniquefeature of the SNi mechanism that accounts

SNi: a form of SN1reaction in which thenucleophile isderived bydecomposition of theleaving group andattacks from thesame face

for the exclusive retention of stereochemistryis that the leaving group undergoes decom-position leading to the production of a nu-cleophile that is held as an ion pair on thesame face as the leaving group. Decomposi-tion of the initial intermediate species and at-tack by the formed nucleophile occur at a ratethat exceeds that of solvent attack and the ionpair reorganization that would be required fornucleophilic attack from the back face. Manyauthors now propose SNi-like mechanisms forretaining glycosyltransferases; this potentiallyconfusing nomenclature demands some his-torical explanation.

In 1980, Sinnott & Jencks (140) re-ported that the solvolysis of glucose deriva-tives in mixtures of ethanol and trifluo-roethanol could, under certain conditions,proceed with retention of anomeric config-uration via a mechanism that they consid-ered “a type of internal return.” The ratio-nale for this description was derived from theproposed requirement for the formation of a


Ann

u. R

ev. B

ioch

em. 2

008.

77. D

ownl

oade

d fr

om a

rjou

rnal

s.an

nual

revi

ews.

org

by U

NIW

ER

SYT

ET

WR

OC

LA

WSK

I on

04/

18/0

8. F

or p

erso

nal u

se o

nly.


hydrogen bond between the departing fluo-ride ion and the attacking nucleophile. Thestereoelectronic context of this species is ex-tremely similar to that displayed in the secondintermediate species of alkyl chlorosulfite de-composition (Figure 8a), although in the lat-ter case the nucleophile was previously a co-valently attached part of the substrate, henceits “internal” nature. Sinnott & Jencks thusdrew the analogy; if one considers F-. . .OR asa single species, then the solvolysis of glycosylfluorides is a type of internal return. They fur-ther suggested that a sufficiently “open” tran-sition state can allow for a nucleophilic pushon the reaction center by the entering sol-vent molecule on either the opposite or thesame face as the leaving group, thereby di-rectly challenging the classical Ingold view ofSN2 displacement reactions involving Waldeninversion.

A mechanism termed SNi-like was subse-quently proposed for glycogen phosphory-lase (141), perhaps unfortunately given thatthe acceptor is external and not internal. Inlight of structural, mutagenesis, and mecha-nistic results with LgtC, the SNi-like mech-anism was proposed (88) and has since beensuggested for the structurally defined retain-ing transferases Kre2 (110), ToxB (109), Extl2(111), Mgs (27), WaaG (132), OtsA (142), andAGT (143) primarily because of the lack of anappropriately positioned nucleophile withintheir active sites and also on account of thelikely interaction between the departing phos-phate and the incoming acceptor nucleophile.One should heed Sinnott’s words on this topic;“the evidence in favor of the internal re-turn mechanism. . .is essentially negative. . . .”(144) This mechanism is often advocated sim-ply because of the lack of supporting evidencefor other mechanisms.

It is useful to draw parallels fromthe decomposition of alkyl chlorosulfitesthrough the solvolysis of glycosyl fluoridesto the retaining glycosyltransferase reaction(Figure 8b). But such a mechanism should,in the absence of evidence to the contrary,follow the rules of Woodward and Hoffman

(145, 146) and should not be drawn as a sin-gle exploded four-center transition state asit frequently is. An enzyme-catalyzed SNi-like mechanism would in all likelihood pro-ceed with the required formation of a short-lived ion pair intermediate with a lifetimelonger than that of a bond vibration (discussedbelow).

Evolutionary Constraints uponRetaining GlycosyltransferaseMechanisms

By contemplating the inherent differences inthe reactivities of the substrates utilized byglycosidases and glycosyltransferases, a ratio-nale can be developed as to why these twoclasses of enzymes may have evolved differ-ing mechanistic strategies for catalyzing gly-cosyl group transfer reactions with net reten-tion of anomeric configuration. Glycosidasescatalyze the hydrolysis of one of the most sta-ble linkages found in nature, the glycosidicbond between two sugars, with the half-livesfor spontaneous hydrolysis being on the or-der of ∼5 million years (76). The stability ofthis linkage, under neutral conditions, is im-parted by the relatively poor leaving groupability (pKa ∼14) of natural glycosidase sub-strates. This stability leads to a large energybarrier to bond cleavage in the first step ofthe glycosidase reaction (Figure 9). It wouldseem logical that retaining glycosidases wouldhave evolved a catalytic mechanism involv-ing an intermediate and, in accordance withthe Hammond postulate, also a transitionstate of the lowest possible free energy. Inother words, by selecting a mechanism witha lower energy intermediate species, catalyticefficiency is easier to achieve because the rate-determining transition state being stabilizedwill also have a lower free energy. By this logic,the Koshland SN2-like double-displacementmechanism, involving the formation of a cova-lently bound glycosyl-enzyme intermediate,would have been preferred over the PhillipsSN1-like mechanism, involving the forma-tion of a higher energy oxocarbenium ion


Ann

u. R

ev. B

ioch

em. 2

008.

77. D

ownl

oade

d fr

om a

rjou

rnal

s.an

nual

revi

ews.

org

by U

NIW

ER

SYT

ET

WR

OC

LA

WSK

I on

04/

18/0

8. F

or p

erso

nal u

se o

nly.


intermediate, during the evolution of the re-taining glycosidase mechanism (Figure 9).

In contrast, glycosyltransferases utilizehigh-energy (substituted) phosphosugardonor substrates. The leaving group abilityfor the donor substrate is much greater tostart with, compared to glycosidase sub-strates, with a pKa of ∼7 for the secondionization of phosphate. The higher freeenergy of the starting material and lowerfree energy of the first transition state makethe barrier for bond cleavage in the first stepof the glycosyltransferase reaction less thanthat of the glycosidase reaction (Figure 9).Because there is not such a large barrier forthe initial bond cleavage step compared tothat for the analogous glycosidase reaction,an SN1-like pathway involving the formationof an intermediate oxocarbenium ion mayhave been accessible, and sufficient to havebeen selected for, during the course of theevolution of the mechanisms of retainingglycosyltransferases (Figure 9).

Precedence for enzymatic mechanisms in-volving SN1-like pathways, and for the in-termediate formation of cationic species, canbe derived from enzymes involved in ter-penoid biosynthesis: isopentenyl diphosphateisomerases, FPP synthetases, and the terpenecyclases. These enzymes use SN1-like mecha-nisms involving the formation of carbocationintermediates stabilized by either aromaticside chains (pi-cationic interactions), the car-bonyl oxygens of main or side chain amides,and/or side chain hydroxyls (147–153).

The chemistry involved in FPPsynthetase-catalyzed reactions is ratheranalogous to that involved in glycosyltrans-ferase catalysis. In both cases, a nucleophilicsubstitution reaction occurs at a carboncenter with a (substituted) phosphate actingas the leaving group. Stereochemically, theyare more like inverting enzymes in that theypreorganize the nucleophilic alkene in aposition to directly trap the carbocation anddo not suffer the same stereochemical imped-iment as do the retaining enzymes. However,analogies between the overall chemistries

being catalyzed provide a foundation for acomparison of the transition states chosento be stabilized and the possibility of aconvergence in mechanism during the courseof evolution. As is observed for glycosyl-transferases, FPP synthetases use a divalentcation and the positively charged side chainsdonated from a conserved Arg/Lys pair tostabilize charge development and facilitatedeparture of the phosphate leaving group(84, 85). Intriguingly, conserved electrostaticinteractions with neutral carbonyl groups,derived from the amides of a main chain(Lys202) and a side chain (Gln241), plus a sidechain hydroxyl (Thr203), are found withinthe FPP synthetase active site and are thoughtto stabilize the positive charge distributedover the C1, C2, and C3 atoms of the cationicintermediate (84, 85). This is similar to theobserved positioning of the carbonyl oxygenof a main chain or a side chain amide in closeproximity (on the β-face) to the anomericreaction center within the active sites of thevast majority of structurally characterizedretaining glycosyltransferases. In both cases,formation of a cationic intermediate is alsostabilized by inherent properties of thesubstrates. In the case of glycosyltransferases,positive charge formation at the anomericcenter of an oxocarbenium ion is stabilized bya lone pair of nonbonding electrons donatedby the endocyclic oxygen of the sugar ring. Inthe case of FPP synthetases, an allylic tertiarycenter stabilizes positive charge developmentat the carbon center that undergoes bondfission. Also, as is proposed for retaining gly-cosyltransferases, FPP synthetases appear toposition the phosphate leaving group in sucha manner that it is suitably positioned to playthe role of the base catalyst required for pro-ton abstraction in the elimination step of thereaction. These described similarities in thechemistries catalyzed and the observed con-servation of structural architecture might wellsuggest that retaining glycosyltransferasesutilize an SN1-like mechanism, analogous tothat of FPP synthetase, which given the needfor interaction between leaving group and


Ann

u. R

ev. B

ioch

em. 2

008.

77. D

ownl

oade

d fr

om a

rjou

rnal

s.an

nual

revi

ews.

org

by U

NIW

ER

SYT

ET

WR

OC

LA

WSK

I on

04/

18/0

8. F

or p

erso

nal u

se o

nly.


b

Cov

alen

t gly

cosy

l-en

zym

e in

term

edia

te(S

N2-

like

path

way

)

GH

pro

du

ctco

mp

lex

GH

sub

stra

teco

mp

lex

Oxo

carb

eniu

mio

n in

term

edia

te(S

N1-

like

path

way

)

O

HO

δ+

δ+

δ+

NucO

NH

2

O

HO

a

GH

pro

du

ctco

mp

lex

GH

sub

stra

teco

mp

lex

O

HO

δ+

δ+

OO N

uc

O

HO

Cov

alen

t gly

cosy

l-en

zym

e in

term

edia

te(S

N2-

like

path

way

)

Oxo

carb

eniu

mio

n in

term

edia

te(S

N1-

like

path

way

)

–

Figu

re9

Com

pari

son

ofth

epo

tent

iald

iffer

ence

sin

the

free

ener

gyba

rrie

rsfo

rS N

1-lik

ean

dS N

2-lik

epa

thw

ays

that

coul

dbe

utili

zed

by(a

)ret

aini

nggl

ycos

idas

es(G

Hs)

and

(b)

reta

inin

ggl

ycos

yltr

ansf

eras

es(G

Ts)

.Bec

ause

glyc

osyl

tran

sfer

ases

utili

zehi

gh-e

nerg

y(s

ubst

itute

d)ph

osph

osug

ardo

nor

subs

trat

esco

ntai

ning

ave

rygo

odle

avin

ggr

oup

(pK

aof

∼2fo

rth

eco

ordi

nate

dsp

ecie

s),t

hefr

eeen

ergy

barr

ier

for

the

first

step

ism

uch

less

than

that

for

the

glyc

osid

ases

that

utili

zea

very

stab

lesu

bstr

ate

cont

aini

nga

poor

leav

ing

grou

p(p

Ka

of∼1

4).T

his

smal

ler

free

ener

gyba

rrie

rfo

rth

efir

stst

epof

the

reac

tion

may

have

led

toth

eS N

1-lik

epa

thw

aybe

ing

suffi

cien

tlyef

fect

ive

tobe

sele

cted

for

duri

ngth

eco

urse

ofth

eev

olut

ion

ofre

tain

ing

glyc

osyl

tran

sfer

ase

mec

hani

sms.

Inad

ditio

n,fo

rgl

ycos

yltr

ansf

eras

es,b

ecau

seof

the

near

equa

lity

inle

avin

ggr

oup

abili

tyfo

rbo

thst

eps,

utili

zatio

nof

asi

dech

ain

carb

oxyl

ate

asa

cata

lytic

nucl

eoph

ile(a

sis

the

case

for

the

vast

maj

ority

ofre

tain

ing

glyc

osid

ases

)in

ado

uble

-dis

plac

emen

tmec

hani

smm

ight

lead

toa

rate

-lim

iting

degl

ycos

ylat

ion

step

.Thi

sw

ould

incr

ease

the

prob

abili

tyof

unw

ante

ddo

nor

subs

trat

ehy

drol

ysis

.


Ann

u. R

ev. B

ioch

em. 2

008.

77. D

ownl

oade

d fr

om a

rjou

rnal

s.an

nual

revi

ews.

org

by U

NIW

ER

SYT

ET

WR

OC

LA

WSK

I on

04/

18/0

8. F

or p

erso

nal u

se o

nly.


attacking nucleophile could be considered atype of internal return.

Concluding Remarks on theMechanism of RetainingGlycosyltransferases

Presently, the catalytic mechanism of retain-ing glycosyltransferases is a topic withoutclear or definitive answers. By analogy tothe well-characterized glycosidases, a double-displacement mechanism involving discretenucleophilic catalysis and the formation of acovalently bound glycosyl-enzyme intermedi-ate would seem possible. Such a mechanismnecessitates an appropriately positioned enzy-matic nucleophile within the active site on theβ-face of the donor substrate in close proxim-ity to the anomeric reaction center. However,candidates for this role are not clear becausein the majority of cases the most suitably po-sitioned functional group is a main chain orside chain amide group. The plausibility of anucleophilic role for these groups is compro-mised by what is observed for the family GT8retaining enzyme LgtC, in which mutation ofthe glutamine (Gln189) whose side chain isso positioned results in an unexpectedly mod-est decrease in observed rates. Furthermore,the finding that a competent glycosyl-enzymeintermediate involving Asp190 is trapped bysubstrate in the Q189E variant of LgtC con-fused issues somewhat but suggests a highlyplastic active site. Alternatively, LgtC maywell use a SN1-like pathway, and the alteredelectrostatic environment in the active siteof the Q189E mutant results in an elec-trophilic oxocarbenium ion being quenchedby a nearby nucleophilic side chain.

A mechanism involving the formation ofan ionic intermediate might be more reason-able for glycosyltransferases than for glyco-side hydrolases, given the greater reactivity oftheir substrates. However, it is also possiblethat some retaining glycosyltransferases usea distinctively SN2-like mechanism involv-ing formation of a covalently bound glycosyl-enzyme intermediate as suggested for the fam-

ily GT6 retaining enzyme α3GalT. Indeed,the presence of a correctly disposed carboxy-late side chain and the catalytic rescue af-forded to mutants at that position by azidesupports such a notion. Overall, it is mostprobable that a mechanistic continuum existsfor this class of enzyme, analogous to that ofnonenzymatic nucleophilic substitution reac-tions for which the SN2 direct displacementand SN1 ionic mechanisms are simply the ex-tremes. Between these two extremes, a con-tinuum of nucleophilic substitution mecha-nisms exists that is best defined using theInternational Union for Pure and AppliedChemistry (IUPAC)-recommended designa-tions for reaction mechanisms (154, 155) (seethe sidebar IUPAC-Recommended Nomen-clature for Reaction Mechanisms). These in-volve ion pair formation and can have the ki-netic properties of an SN2 process yet produce

IUPAC-RECOMMENDED NOMENCLATUREFOR REACTION MECHANISMS

In 1989, following several years of vigorous debate, the IUPACcommission on physical organic chemistry described a systemfor the systematic description of reaction mechanisms (154,159). This system was proposed to replace the Ingold systemthat served to define the type of transformation and the empir-ically observed molecularity (e.g., SN2) but not the details ofthe mechanism (e.g., concerted versus step wise). Although theuse of this system may appear to be a purely semantic exercise,its benefit is clear in light of the frequent unnecessary contro-versies that arise during written discussions of reaction mecha-nisms, resulting from ambiguities of the meaning of the Ingolddefinitions. This argument is particularly applicable to the dis-cussion of the mechanism of retaining glycosyltransferases.Using the IUPAC recommended nomenclature, the SNi-likemechanism described for retaining glycosyltransferases wouldbe defined as DN

∗ANss, where the asterisk represents the for-mation of a short-lived intermediate and “ss” represents theformation of a solvent-separated ion pair (Figure 10). Al-though the term “solvent separated” seems unusual in the con-text of an enzymatic mechanism, it is required to differentiateit from an “intimate” ion pair that would not allow for attackby an external acceptor from the front face.


Ann

u. R

ev. B

ioch

em. 2

008.

77. D

ownl

oade

d fr

om a

rjou

rnal

s.an

nual

revi

ews.

org

by U

NIW

ER

SYT

ET

WR

OC

LA

WSK

I on

04/

18/0

8. F

or p

erso

nal u

se o

nly.


Figure 10Proposed DN

∗ANss ion pair mechanism for retaining glycosyltransferases.

stereochemical outcomes consistent with anSN1 reaction (and vice versa) (156–158).

The mechanism that seems likely formost retaining glycosyltransferases, includingLgtC, is one involving a short-lived ion pairintermediate that requires a back-side nucle-ophilic “push” (without discrete covalent in-teraction) for formation (Figure 10). This ionpair could collapse to give back starting ma-terial or, following a slight shift in the active-site positioning of the cation, could undergo

front-side attack by an incoming nucleophile,leading to a product with retained stereo-chemistry. Indeed, consideration of the reac-tivities of the substrates in question, the lack ofconserved structural architecture among re-taining glycosyltransferase β-face active-siteregions, and the biological precedents for suchintermediates in related cationic mechanismssuggest that this is the most probable mech-anism for the majority of retaining glycosyl-transferases.


Ann

u. R

ev. B

ioch

em. 2

008.

77. D

ownl

oade

d fr

om a

rjou

rnal

s.an

nual

revi

ews.

org

by U

NIW

ER

SYT

ET

WR

OC

LA

WSK

I on

04/

18/0

8. F

or p

erso

nal u

se o

nly.


SUMMARY POINTS

1. Glycosyltransferases catalyze group transfer reactions with either net inversion orretention of stereochemistry at the anomeric reaction with respect to the (substituted)phosphosugar donor substrate, thereby necessitating differing catalytic mechanismsfor these two classes of enzymes.

2. Thus far, the 3-D structures of glycosyltransferases have revealed two major struc-tural folds, with others just now being uncovered. The GT-A and GT-B folds,which incorporate a pair of Rossmann folds in each case, are found for nucleosidephosphosugar-dependent glycosyltransferases. Lipid phosphosugar-dependent gly-cosyltransferases appear to adopt different folds, as exemplified by the lysozyme-likefold reported for a peptidoglycan synthase and the novel fold recently found for anoligosaccharyltransferase.

3. The mechanism of inverting glycosyltransferases is that of a straightforward SN2-likereaction facilitated by an enzymatic base catalyst and by Lewis acid activation of thedeparting (substituted) phosphate leaving group.

4. The mechanism of retaining glycosyltransferases remains less clear. Although somemembers of this class may utilize Koshland’s double-displacement mechanism involv-ing the formation of a covalently bound glycosyl-enzyme intermediate, a mechanisminvolving the formation of a short-lived ion pair intermediate seems likely for themajority of the members of this class of enzyme.

5. There is no correlation of overall fold with catalytic mechanism. Inverting and re-taining enzymes are known with both GT-A and GT-B topologies.

FUTURE ISSUES

1. Does the predicted GT-C fold have any predictive relevance beyond indicating thepresence of a large trans-membrane component? First indications from the very re-cently solved structure of the oligosaccharyltransferase STT3 suggest it may not.

2. Will a whole new series of folds for nonnucleoside phosphosugar-dependent glycosyl-transferases be discovered? In the absence of the constraints of (nucleotide-binding)Rossmann folds, the structural diversity may well rival that of the glycosidases.

3. When is a transferase a class of hydrolase? Both glycoside hydrolases and transferasescatalyze glycosyl group transfer, and whether an enzyme is classified as one or the otherfollows few clear rules. Classification is sometimes based upon thermodynamics, oftenupon sequence, and occasionally upon structural homologies, and it is frequentlyinfluenced by historical perspective. For example, family GT51 is demonstrably aglycosyltransferase but could, arguably, be reclassified to a glycoside hydrolase familygiven its structural similarities to hydrolyzing lysozymes. This can occasionally be aconfusing area for the nonspecialist.

4. The overall folds of multiple “orphan” families of glycosyltransferases remain to bedetermined.


Ann

u. R

ev. B

ioch

em. 2

008.

77. D

ownl

oade

d fr

om a

rjou

rnal

s.an

nual

revi

ews.

org

by U

NIW

ER

SYT

ET

WR

OC

LA

WSK

I on

04/

18/0

8. F

or p

erso

nal u

se o

nly.


5. The catalytic mechanism of retaining glycosyltransferases remains ill defined.Whether some members (e.g., family GT6) utilize a double-displacement mecha-nism and others an ion pair mechanism remains to be determined.

6. The structure and mechanism of the major polymerizing glycosyltransferases, suchas those involved in chitin, cellulose, and hyaluronan biosynthesis, need to be de-termined as well as a resolution of the continuing controversies as to whether theUDP-monosaccharide is the glycosyl donor (nonreducing end addition) or the UDP-growing chain (reducing end addition).

7. Many, probably all, glycosyltransferases display conformational changes upon ligandbinding and catalysis. For most enzymes the nature and extent of these changes hasnot been well-defined.

8. Perhaps the greatest challenge in the field is the functional characterization of glyco-syltransferases. There are over 33,000 open reading frames known that encode thisclass of enzyme (as of January 2008), yet the donor and acceptor specificity for thevast majority (>95%) is not known.

DISCLOSURE STATEMENT

The authors are not aware of any biases that might be perceived as affecting the objectivity ofthis review.

ACKNOWLEDGMENTS

L.L.L is the receipient of a Natural Science and Engineering Reseach Council (NSERC) doc-toral postgraduate scholarship and a Michael Smith Foundation for Health Research (MSFHR)senior graduate fellowship. G.J.D. is a Royal Society-Wolfson Research Merit Award recipient.B.H. thanks the Centre National de la Recherche Scientifique for supporting the visit of S.G.W.to France. S.G.W. thanks the Natural Sciences and Engineering Research Council of Canadaand the Canadian Institutes for Health Research for continuing support.

LITERATURE CITED

1. Rademacher TW, Parekh RB, Dwek RA. 1988. Annu. Rev. Biochem. 57:785–8382. Davies GS, Withers SG. 1998. In Comprehensive Biological Catalysis, ed. ML Sinnott,

pp. 119–208. London: Academic3. Zechel DL, Withers SG. 2000. Acc. Chem. Res. 33:11–184. Campbell JA, Davies GJ, Bulone V, Henrissat B. 1997. Biochem. J. 326:929–395. Coutinho PM, Deleury E, Davies GJ, Henrissat B. 2003. J. Mol. Biol. 328:307–176. Bourne Y, Henrissat B. 2001. Curr. Opin. Struc. Biol. 11:593–6007. Unligil UM, Rini JM. 2000. Curr. Opin. Struc. Biol. 10:510–178. Hu YN, Walker S. 2002. Chem. Biol. 9:1287–969. Charnock SJ, Davies GJ. 1999. Biochemistry 38:6380–85

10. Breton C, Imberty A. 1999. Curr. Opin. Struc. Biol. 9:563–7111. Brown K, Pompeo F, Dixon S, Mengin-Lecreulx D, Cambillau C, Bourne Y. 1999. EMBO

J. 18:4096–107


Ann

u. R

ev. B

ioch

em. 2

008.

77. D

ownl

oade

d fr

om a

rjou

rnal

s.an

nual

revi

ews.

org

by U

NIW

ER

SYT

ET

WR

OC

LA

WSK

I on

04/

18/0

8. F

or p

erso

nal u

se o

nly.


12. Chiu CPC, Watts AG, Lairson LL, Gilbert M, Lim D, et al. 2004. Nat. Struct. Mol. Biol.11:163–70

13. Chiu CPC, Lairson LL, Gilbert M, Wakarchuk WW, Withers SG, Strynadka NCJ. 2007.Biochemistry 46:7196–204

14. Wiggins CAR, Munro S. 1998. Proc. Natl. Acad. Sci. USA 95:7945–5015. Breton C, Bettler E, Joziasse DH, Geremia RA, Imberty A. 1998. J. Biochem. 123:1000–916. Pak JE, Arnoux P, Zhou SH, Sivarajah P, Satkunarajah M, et al. 2006. J. Biol. Chem.

281:26693–70117. Vrielink A, Ruger W, Driessen HPC, Freemont PS. 1994. EMBO J. 13:3413–2218. Campbell RE, Mosimann SC, Tanner ME, Strynadka NCJ. 2000. Biochemistry 39:14993–

500119. Liu J, Mushegian A. 2003. Protein Sci. 12:1418–3120. Strahlbolsinger S, Immervoll T, Deutzmann R, Tanner W. 1993. Proc. Natl. Acad. Sci.

USA 90:8164–6821. Takahashi M, Inoue N, Ohishi K, Maeda Y, Nakamura N, et al. 1996. EMBO J. 15:4254–

6122. Maeda Y, Watanabe R, Harris CL, Hong JJ, Ohishi K, et al. 2001. EMBO J. 20:250–6123. Kikuchi N, Kwon YD, Gotoh M, Narimatsu H. 2003. Biochem. Biophys. Res. Commun.

310:574–7923a. Igura M, Maita N, Kamishikiryo J, Yamada M, Obita T, et al. 2007. EMBO J. 27:234–4324. Hidaka M, Honda Y, Kitaoka M, Nirasawa S, Hayashi K, et al. 2004. Structure 12:937–4725. Lovering AL, de Castro LH, Lim D, Strynadka NCJ. 2007. Science 315:1402–526. Yuan YQ, Barrett D, Zhang Y, Kahne D, Sliz P, Walker S. 2007. Proc. Natl. Acad. Sci.

USA 104:5348–5327. Flint J, Taylor E, Yang M, Bolam DN, Tailford LE, et al. 2005. Nat. Struct. Mol. Biol.

12:608–1428. Tarbouriech N, Charnock SJ, Davies GJ. 2001. J. Mol. Biol. 314:655–6129. Gastinel LN, Cambillau C, Bourne Y. 1999. EMBO J. 18:3546–5730. Ramakrishnan B, Qasba PK. 2001. J. Mol. Biol. 310:205–1831. Ramakrishnan B, Shah PS, Qasba PK. 2001. J. Biol. Chem. 276:37665–7132. Ramakrishnan B, Balaji PV, Qasba PK. 2002. J. Mol. Biol. 318:491–50233. Ramasamy V, Ramakrishnan B, Boeggeman E, Ratner DM, Seeberger PH, Qasba PK.

2005. J. Mol. Biol. 353:53–6734. Ramakrishnan B, Ramasamy V, Qasba PK. 2006. J. Mol. Biol. 357:1619–3335. Ramakrishnan B, Qasba PK. 2002. J. Biol. Chem. 277:20833–3936. Unligil UM, Zhou SH, Yuwaraj S, Sarkar M, Schachter H, Rini JM. 2000. EMBO J.

19:5269–8037. Gordon RD, Sivarajah P, Satkunarajah M, Ma D, Tarling CA, et al. 2006. J. Mol. Biol.

360:67–7938. Pedersen LC, Tsuchida K, Kitagawa H, Sugahara K, Darden TA, Negishi M. 2000. J.

Biol. Chem. 275:34580–8539. Pedersen LC, Darden TA, Negishi M. 2002. J. Biol. Chem. 277:21869–7340. Jinek M, Chen YW, Clausen H, Cohen SM, Conti E. 2006. Nat. Struct. Mol. Biol. 13:945–

4641. Ohtsubo K, Imajo S, Ishiguro M, Nakatani T, Oka S, Kawasaki T. 2000. J. Biochem.

128:283–9142. Kakuda S, Shiba T, Ishiguro M, Tagawa H, Oka S, et al. 2004. J. Biol. Chem. 279:22693–

703


Ann

u. R

ev. B

ioch

em. 2

008.

77. D

ownl

oade

d fr

om a

rjou

rnal

s.an

nual

revi

ews.

org

by U

NIW

ER

SYT

ET

WR

OC

LA

WSK

I on

04/

18/0

8. F

or p

erso

nal u

se o

nly.


43. Garinot-Schneider C, Lellouch AC, Geremia RA. 2000. J. Biol. Chem. 275:31407–1344. Kozmon S, Tvaroska I. 2006. J. Am. Chem. Soc. 128:16921–2745. Koyama M, Helbert W, Imai T, Sugiyama J, Henrissat B. 1997. Proc. Natl. Acad. Sci. USA

94:9091–9546. Kamst E, Bakkers J, Quaedvlieg NEM, Pilling J, Kijne JW, et al. 1999. Biochemistry

38:4045–5247. Chang R, Yeager AR, Finney NS. 2003. Org. Biomol. Chem. 1:39–4148. DeAngelis PL. 1999. J. Biol. Chem. 274:26557–6249. Tlapak-Simmons VL, Baron CA, Gotschall R, Haque D, Canfield WM, Weigel PH.

2005. J. Biol. Chem. 280:13012–1850. Kornberg SR, Zimmerman SB, Kornberg A. 1961. J. Biol. Chem. 236:1487–9351. Morera S, Lariviere L, Kurzeck J, Aschke-Sonnenborn U, Freemont PS, et al. 2001. J.

Mol. Biol. 311:569–7752. Lariviere L, Morera S. 2002. J. Mol. Biol. 324:483–9053. Lariviere L, Gueguen-Chaignon V, Morera S. 2003. J. Mol. Biol. 330:1077–8654. Mulichak AM, Losey HC, Lu W, Wawrzak Z, Walsh CT, Garavito RM. 2003. Proc. Natl.

Acad. Sci. USA 100:9238–4355. Mulichak AM, Losey HC, Walsh CT, Garavito RM. 2001. Structure 9:547–5756. Mulichak AM, Lu W, Losey HC, Walsh CT, Garavito RM. 2004. Biochemistry 43:5170–

8057. Shao H, He XZ, Achnine L, Blount JW, Dixon RA, Wang XQ. 2005. Plant Cell 17:3141–

5458. Offen W, Martinez-Fleites C, Yang M, Kiat-Lim E, Davis BG, et al. 2006. EMBO J.

25:1396–40559. Bolam DN, Roberts S, Proctor MR, Turkenburg JP, Dodson EJ, et al. 2007. Proc. Natl.

Acad. Sci. USA 104:5336–4160. Miley MJ, Zielinska AK, Keenan JE, Bratton SM, Radominska-Pandya A, Redinbo MR.

2007. J. Mol. Biol. 369:498–51160a. Brazier-Hicks M, Offen WA, Gershater MC, Revett TJ, Lim EK, et al. 2007. Proc. Natl.

Acad. Sci. USA 104:20238–4361. Grizot S, Salem M, Vongsouthi V, Durand L, Moreau F, et al. 2006. J. Mol. Biol. 363:383–

9462. Sun HY, Lin SW, Ko TP, Pan JF, Liu CL, et al. 2007. J. Biol. Chem. 282:9973–8263. Ihara H, Ikeda Y, Toma S, Wang XC, Suzuki T, et al. 2007. Glycobiology 17:455–6664. Ha S, Walker D, Shi YG, Walker S. 2000. Protein Sci. 9:1045–5265. Hu YN, Chen L, Ha S, Gross B, Falcone B, et al. 2003. Proc. Natl. Acad. Sci. USA

100:845–4966. Yu H, Chokhawala H, Karpel R, Yu H, Wu BY, et al. 2005. J. Am. Chem. Soc. 127:17618–

1967. Zhang CS, Griffith BR, Fu Q, Albermann C, Fu X, et al. 2006. Science 313:1291–9468. Ni LS, Sung MC, Yu H, Chokhawala H, Chen X, Fisher AJ. 2006. Biochemistry 45:2139–

4869. Ni L, Chokhawala HA, Cao H, Henning R, Ng L, et al. 2007. Biochemistry 46:6288–9870. Yip VLY, Varrot A, Davies GJ, Rajan SS, Yang XJ, et al. 2004. J. Am. Chem. Soc. 126:8354–

5571. Liu QYP, Sulzenbacher G, Yuan HP, Bennett EP, Pietz G, et al. 2007. Nat. Biotechnol.

25:454–6472. Mark BL, James MNG. 2002. Can. J. Chem. 80:1064–74


Ann

u. R

ev. B

ioch

em. 2

008.

77. D

ownl

oade

d fr

om a

rjou

rnal

s.an

nual

revi

ews.

org

by U

NIW

ER

SYT

ET

WR

OC

LA

WSK

I on

04/

18/0

8. F

or p

erso

nal u

se o

nly.


73. Koshland DE. 1953. Biol. Rev. Camb. Phil. Soc. 28:416–3674. Phillips DC. 1966. Sci. Am. 215:78–9075. Phillips DC. 1967. Proc. Natl. Acad. Sci. USA 57:484–9576. Wolfenden R, Lu XD, Young G. 1998. J. Am. Chem. Soc. 120:6814–1577. Fersht A. 1999. Structure and Mechanism in Protein Science: A Guide to Enzyme Catalysis and

Protein Folding. New York: Freeman78. Sinnott ML, Souchard IJ. 1973. Biochem. J. 133:89–9879. Kempton JB, Withers SG. 1992. Biochemistry 31:9961–6980. Vocadlo DJ, Wicki J, Rupitz K, Withers SG. 2002. Biochemistry 41:9727–3581. Huang X, Tanaka KSE, Bennet AJ. 1997. J. Am. Chem. Soc. 119:11147–5482. Berti PJ, Tanaka KSE. 2002. In Advances in Physical Organic Chemistry, Vol 37. pp. 239–

314, ed. T Tidwell, J Richard. Oxford, UK: Elsevier83. Vocadlo DJ, Davies GJ, Laine R, Withers SG. 2001. Nature 412:835–3884. Hosfield DJ, Zhang YM, Dougan DR, Broun A, Tari LW, et al. 2004. J. Biol. Chem.

279:8526–2985. Gabelli SB, McLellan JS, Montalvetti A, Oldfield E, Docampo R, Amzel LM. 2006.

Proteins 62:80–886. Withers SG, Aebersold R. 1995. Protein Sci. 4:361–7287. Ly HD, Lougheed B, Wakarchuk WW, Withers SG. 2002. Biochemistry 41:5075–8588. Persson K, Ly HD, Dieckelmann M, Wakarchuk WW, Withers SG, Strynadka NCJ.

2001. Nat. Struct. Biol. 8:166–7589. Lairson LL, Chiu CPC, Ly HD, He SM, Wakarchuk WW, et al. 2004. J. Biol. Chem.

279:28339–4490. Gibbons BJ, Roach PJ, Hurley TD. 2002. J. Mol. Biol. 319:463–7791. Gastinel LN, Bignon C, Misra AK, Hindsgaul O, Shaper JH, Joziasse DH. 2001. EMBO

J. 20:638–4992. Zhang YN, Swaminathan GJ, Deshpande A, Boix E, Natesh R, et al. 2003. Biochemistry

42:13512–2193. Jamaluddin H, Tumbale P, Withers SG, Acharya KR, Brew K. 2007. J. Mol. Biol.

369:1270–8194. Monegal A, Planas A. 2006. J. Am. Chem. Soc. 128:16030–3195. Watkins WM, Morgan WTJ. 1957. Nature 180:1038–4196. Yamamoto F, Clausen H, White T, Marken J, Hakomori SI. 1990. Nature 345:229–3397. Patenaude SI, Seto NOL, Borisova SN, Szpacenko A, Marcus SL, et al. 2002. Nat. Struct.

Biol. 9:685–9098. Lee HJ, Barry CH, Borisova SN, Seto NOL, Zheng RXB, et al. 2005. J. Biol. Chem.

280:525–2999. Nguyen HP, Seto NOL, Cai Y, Leinala EK, Borisova SN, et al. 2003. J. Biol. Chem.

278:49191–95100. Letts JA, Rose NL, Fang YR, Barry CH, Borisova SN, et al. 2006. J. Biol. Chem. 281:3625–

32101. Persson M, Letts JA, Hosseini-Maaf B, Borisova SN, Palcic MM, et al. 2007. J. Biol.

Chem. 282:9564–70102. Hosseini-Maaf B, Letts JA, Persson M, Smart E, LePennec PY, et al. 2007. Transfusion

47:864–75103. Ten Hagen KG, Tetaert D, Hagen FK, Richet C, Beres TM, et al. 1999. J. Biol. Chem.

274:27867–74104. Ten Hagen KG, Bedi GS, Tetaert D, Kingsley PD, Hagen FK, et al. 2001. J. Biol. Chem.

276:17395–404


Ann

u. R

ev. B

ioch

em. 2

008.

77. D

ownl

oade

d fr

om a

rjou

rnal

s.an

nual

revi

ews.

org

by U

NIW

ER

SYT

ET

WR

OC

LA

WSK

I on

04/

18/0

8. F

or p

erso

nal u

se o

nly.


105. Fritz TA, Hurley JH, Trinh LB, Shiloach J, Tabak LA. 2004. Proc. Natl. Acad. Sci. USA101:15307–12

106. Kubota T, Shiba T, Sugioka S, Furukawa S, Sawaki H, et al. 2006. J. Mol. Biol. 359:708–27107. Fritz TA, Raman J, Tabak LA. 2006. J. Biol. Chem. 281:8613–19108. Just I, Selzer J, Wilm M, Voneichelstreiber C, Mann M, Aktories K. 1995. Nature 375:500–

3109. Reinert DJ, Jank T, Aktories K, Schulz GE. 2005. J. Mol. Biol. 351:973–81110. Lobsanov YD, Romero PA, Sleno B, Yu BM, Yip P, et al. 2004. J. Biol. Chem. 279:17921–

31111. Pedersen LC, Dong J, Taniguchi F, Kitagawa H, Krahn JM, et al. 2003. J. Biol. Chem.

278:14420–28112. Green AA, Cori GT, Cori CF. 1942. J. Biol. Chem. 142:447–48113. Fisher EH, Graves DJ, Snyder Crittended ER, Krebs EG. 1959. J. Biol. Chem. 1959:1698–

704114. Fletterick RJ, Sprang SR. 1982. Acc. Chem. Res. 15:361–69115. Raibaud O, Schwartz M. 1984. Annu. Rev. Genet. 18:173–206116. Watson KA, Mitchell EP, Johnson LN, Son JC, Bichard CJF, et al. 1994. Biochemistry

33:5745–58117. Mitchell EP, Withers SG, Ermert P, Vasella AT, Garman EF, et al. 1996. Biochemistry

35:7341–55118. O’Reilly M, Watson KA, Schinzel R, Palm D, Johnson LN. 1997. Nat. Struct. Biol. 4:405–

12119. O’Reilly M, Watson KA, Johnson LN. 1999. Biochemistry 38:5337–45120. Geremia S, Campagnolo M, Schinzel R, Johnson LN. 2002. J. Mol. Biol. 322:413–23121. Watson KA, McCleverty C, Geremia S, Cottaz S, Driguez H, Johnson LN. 1999. EMBO

J. 18:4619–32122. Lin K, Rath VL, Dai SC, Fletterick RJ, Hwang PK. 1996. Science 273:1539–41123. Lukacs CM, Oikonomakos NG, Crowther RL, Hong LN, Kammlott RU, et al. 2006.

Proteins 63:1123–26124. Rath VL, Ammirati M, LeMotte PK, Fennell KF, Mansour MN, et al. 2000. Mol. Cell

6:139–48125. Ekstrom JL, Pauly TA, Carty MD, Soeller WC, Culp J, et al. 2002. Chem. Biol. 9:915–24126. Wright SW, Rath VL, Genereux PE, Hageman DL, Levy CB, et al. 2005. Bioorg. Med.

Chem. Lett. 15:459–65127. Klabunde T, Wendt KU, Kadereit D, Brachvogel V, Burger HJ, et al. 2005. J. Med. Chem.

48:6178–93128. Withers SG, Shechosky S, Madsen NB. 1982. Biochem. Biophys. Res. Commun. 108:322–28129. Stirtan WG, Withers SG. 1996. Biochemistry 35:15057–64130. Withers SG, Madsen NB, Sykes BD, Takagi M, Shimomura S, Fukui T. 1981. J. Biol.

Chem. 256:10759–62131. Withers SG, Wakarchuk WW, Strynadka NCJ. 2002. Chem. Biol. 9:1270–73132. Martinez-Fleites C, Proctor M, Roberts S, Bolam DN, Gilbert HJ, Davies GJ. 2006.

Chem. Biol. 13:1143–52133. Gibson RP, Tarling CA, Roberts S, Withers SG, Davies GJ. 2004. J. Biol. Chem. 279:1950–

55134. Lariviere L, Sommer N, Morera S. 2005. J. Mol. Biol. 352:139–50135. Buschiazzo A, Ugalde JE, Guerin ME, Shepard W, Ugalde RA, Alzari PM. 2004. EMBO

J. 23:3196–205


Ann

u. R

ev. B

ioch

em. 2

008.

77. D

ownl

oade

d fr

om a

rjou

rnal

s.an

nual

revi

ews.

org

by U

NIW

ER

SYT

ET

WR

OC

LA

WSK

I on

04/

18/0

8. F

or p

erso

nal u

se o

nly.


136. Horcajada C, Guinovart JJ, Fita I, Ferrer JC. 2006. J. Biol. Chem. 281:2923–31137. Cowdrey WA, Hughes ED, Ingold CK, Masterman S, Scott AD. 1937. J. Chem. Soc.

1253–71138. Hughes ED, Ingold CK, Whitfield IC. 1941. Nature 147:206–7139. Lewis ES, Boozer CE. 1952. J. Am. Chem. Soc. 74:308–11140. Sinnott ML, Jencks WP. 1980. J. Am. Chem. Soc. 102:2026–32141. Klein HW, Im MJ, Palm D. 1986. Eur. J. Biochem. 157:107–14142. Gibson RP, Turkenburg JP, Charnock SJ, Lloyd R, Davies GJ. 2002. Chem. Biol. 9:1337–

46143. Sommer N, Depping R, Piotrowski M, Ruger W. 2004. Biochem. Biophys. Res. Commun.

323:809–15144. Sinnott ML. 1990. Chem. Rev. 90:1171–202145. Schreiner PR, Schleyer PvR, Hill RK. 1994. J. Org. Chem. 59:1849–54146. Moss RA, Fu XL, Sauers RR. 2007. J. Phys. Org. Chem. 20:1–10147. Christianson DW. 2006. Chem. Rev. 106:3412–42148. Muehlbacher M, Poulter CD. 1985. J. Am. Chem. Soc. 107:8307–8149. Muehlbacher M, Poulter CD. 1988. Biochemistry 27:7315–28150. Reardon JE, Abeles RH. 1986. Biochemistry 25:5609–16151. Durbecq V, Sainz G, Oudjama Y, Clantin B, Bompard-Gilles C, et al. 2001. EMBO J.

20:1530–37152. Starks CM, Back KW, Chappell J, Noel JP. 1997. Science 277:1815–20153. Shishova EY, Di Costanzo L, Cane DE, Christianson DW. 2007. Biochemistry 46:1941–51154. Guthrie RD, Jencks WP. 1989. Acc. Chem. Res. 22:343–49155. Jencks WP. 1996. J. Phys. Org. Chem. 9:337–40156. Winstein S, Clippinger E, Fainberg AH, Heck R, Robinson GC. 1956. J. Am. Chem. Soc.

78:328–35157. Jencks WP. 1980. Acc. Chem. Res. 13:161–69158. Carey FA, Sundberg RJ. 1990. Advanced Organic Chemistry. Part A: Structure and Mecha-

nisms. New York: Plenum159. Comm. Phys. Org. Chem. I. 1989. Pure Appl. Chem. 61:23–56

RELATED RESOURCES

Univ. Aix-Marseille I & II. 2007. CAZy database. http://www.cazy.orgConsortium for Functional Glycomics/Nature. 2007. http://www.functionalglycomics.org/


Ann

u. R

ev. B

ioch

em. 2

008.

77. D

ownl

oade

d fr

om a

rjou

rnal

s.an

nual

revi

ews.

org

by U

NIW

ER

SYT

ET

WR

OC

LA

WSK

I on

04/

18/0

8. F

or p

erso

nal u

se o

nly.

E V I E W R S April 14 , 2008. (Minor changes may …...E V I E W S I N A D V A N C E...

Documents

Transcript of E V I E W R S April 14 , 2008. (Minor changes may …...E V I E W S I N A D V A N C E...